repo stringclasses 15
values | fix_commit stringlengths 40 40 | buggy_commit stringlengths 40 40 | message stringlengths 3 64.3k | files listlengths 1 300 | timestamp timestamp[s]date 2013-03-13 20:45:00 2026-04-11 07:48:46 |
|---|---|---|---|---|---|
facebook/react | fdbfac8f8216839520d093513908918de591468f | 746c1dbed581ad128f0ea22b8fe20267391e1c2f | Repro for bug with memo vars at wrong block scope
Minimal(ish) repro of a bug we saw internally, where an output of a nested
reactive scope is defined at the wrong block scope, and so later references to
that value are invalid.
The simplified structure is:
```
scope0 inputs=[] outputs=[] {
scope1 inputs=[] ... | [
{
"path": "compiler/forget/src/__tests__/fixtures/compiler/_bug.inner-memo-value-not-promoted-to-outer-scope-dynamic.expect.md",
"patch": "@@ -0,0 +1,127 @@\n+\n+## Input\n+\n+```javascript\n+function Component(props) {\n+ const item = useFragment(FRAGMENT, props.item);\n+ useFreeze(item);\n+\n+ const co... | 2023-04-18T23:10:00 |
electron/electron | e9a43be9beb91a4cb7af04242fe9e27955aa675c | 254dbd7400797a0c65b6c3711b1008b9bfb541f5 | docs: add IPC doc (#32059)
* docs: add IPC doc
* fix: use "string" primitive
* use 'string' ipcrenderer
* use "number" primitive
* Update docs/tutorial/ipc.md
Co-authored-by: Jeremy Rose <nornagon@nornagon.net>
* Update docs/tutorial/ipc.md
Co-authored-by: Jeremy Rose <nornagon@nornagon.net>
* ... | [
{
"path": "docs/api/ipc-main.md",
"patch": "@@ -1,3 +1,10 @@\n+---\n+title: \"ipcMain\"\n+description: \"Communicate asynchronously from the main process to renderer processes.\"\n+slug: ipc-main\n+hide_title: false\n+---\n+\n # ipcMain\n \n > Communicate asynchronously from the main process to renderer pro... | 2022-02-09T16:00:05 |
huggingface/transformers | 0b2900dd7ae8c6024f820db777830415bb70d44e | 281eeef1bbe7442646892059f2d13ff27c03f4a0 | Fix-release-ubild (#43773)
* fix sam hq
* and sam
* urllib 3 | [
{
"path": "src/transformers/models/sam/processing_sam.py",
"patch": "@@ -61,9 +61,9 @@ class SamImagesKwargs(ImagesKwargs, total=False):\n \"\"\"\n \n segmentation_maps: ImageInput | None\n- input_points: NestedList | torch.Tensor | None\n- input_labels: NestedList | int | torch.Tensor | None\... | 2026-02-06T08:56:50 |
ollama/ollama | ec46f3286c7a6a4dc7bcfb373d63ca909952ea4a | 45a13b1dec1ccc1771cd28af826d603d9c885fd9 | engine: error on embeddings; not currently implemented | [
{
"path": "runner/ollamarunner/runner.go",
"patch": "@@ -691,65 +691,6 @@ type EmbeddingResponse struct {\n \tEmbedding []float32 `json:\"embedding\"`\n }\n \n-func (s *Server) embeddings(w http.ResponseWriter, r *http.Request) {\n-\tvar req EmbeddingRequest\n-\tif err := json.NewDecoder(r.Body).Decode(&req... | 2025-03-13T18:22:19 |
ggml-org/llama.cpp | db9d8aa428012cc5593e18635d4c3c54095f5138 | ccb87fa3ee1961ec915f77cb447706f471dca6a5 | ggml-cuda: native bf16 flash attention for vec kernel (#20525)
* ggml-cuda: native bf16 flash attention for vec and tile kernels
mma kernel still converts bf16 to fp16 before launch, native mma bf16 todo
* ggml-cuda: address code owner review feedback
reverted tile kernel changes to avoid larger refactor
* fix ci ... | [
{
"path": "ggml/src/ggml-cuda/CMakeLists.txt",
"patch": "@@ -116,12 +116,11 @@ if (CUDAToolkit_FOUND)\n list(APPEND GGML_SOURCES_CUDA ${SRCS})\n add_compile_definitions(GGML_CUDA_FA_ALL_QUANTS)\n else()\n- file(GLOB SRCS \"template-instances/fattn-vec*q4_0-q4_0.cu\")\n- l... | 2026-03-22T10:05:51 |
facebook/react | 746c1dbed581ad128f0ea22b8fe20267391e1c2f | de89ed32a2b649ad542a4f3387312aa3c171d3a7 | Add common (Relay) hooks to playground environment
This aligns the playground configuration with our internal compiler
configuration to make it easier to repro compilation issues on playground. There
is a bug that doesn't repro right now and i suspect it's because of different
hooks being configured.
Test plan: ... | [
{
"path": "compiler/forget/packages/playground/components/Editor/index.tsx",
"patch": "@@ -9,9 +9,12 @@ import { parse, ParserPlugin } from \"@babel/parser\";\n import traverse, { NodePath } from \"@babel/traverse\";\n import * as t from \"@babel/types\";\n import {\n+ Effect,\n+ Hook,\n printHIR,\n p... | 2023-04-18T22:29:17 |
huggingface/transformers | 438856e44ae31d188eca096e5ba151254803fb1a | 393b4b3d28e29b4b05b19b4b7f3242a7fc893637 | Fix `convert_rope_params_to_dict` so it uses `rope_theta` from the config (#43766)
* Fix `convert_rope_params_to_dict` so it uses config attrs
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
* review comment
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
---------
... | [
{
"path": "src/transformers/modeling_rope_utils.py",
"patch": "@@ -634,11 +634,17 @@ def convert_rope_params_to_dict(self, ignore_keys_at_rope_validation: set | None\n self.rope_parameters = rope_scaling or self.rope_parameters\n self.rope_parameters = self.rope_parameters if self.rope_param... | 2026-02-06T07:31:41 |
electron/electron | ac1d426c518d57ead45f8634513e7da1c78273e0 | 841d223b3b04ce247eacaa2fc01a6588ae9b01ae | fix: pointer lock escape handling (#32369) | [
{
"path": "shell/browser/api/electron_api_web_contents.cc",
"patch": "@@ -1244,6 +1244,9 @@ bool WebContents::PlatformHandleKeyboardEvent(\n content::KeyboardEventProcessingResult WebContents::PreHandleKeyboardEvent(\n content::WebContents* source,\n const content::NativeWebKeyboardEvent& event) {\n... | 2022-02-09T09:40:50 |
ollama/ollama | 5e2e0b46b1b4ae7145011642631a7b507a70fb03 | 45a13b1dec1ccc1771cd28af826d603d9c885fd9 | fix: error if image requested without vision model | [
{
"path": "model/model.go",
"patch": "@@ -22,6 +22,8 @@ import (\n \t\"github.com/ollama/ollama/model/input\"\n )\n \n+var ErrNoVisionModel = errors.New(\"vision model not found\")\n+\n // Model implements a specific model architecture, defining the forward pass and any model-specific configuration\n type M... | 2025-03-13T17:37:31 |
ggml-org/llama.cpp | 990e4d96980d0b016a2b07049cc9031642fb9903 | 212f4521b013a3eeb79e15df7ca07a5329d39d4b | common/grammar: fix grammar parsing issues to prevent stack overflow and hangs (#18604)
* grammar: add test case for nullable symbol loop
Reproduce stack overflow (or OOM) with ( [x]* )* found while adding
GBNF support to ripgrep-edit.
llama-server reproducer:
curl \
-X POST \
-d '{
"messages": [{ "role": "... | [
{
"path": "src/llama-grammar.cpp",
"patch": "@@ -7,6 +7,7 @@\n #include <cmath>\n #include <algorithm>\n #include <cstdint>\n+#include <set>\n #include <stdexcept>\n \n #define MAX_REPETITION_THRESHOLD 2000\n@@ -454,6 +455,7 @@ const char * llama_grammar_parser::parse_sequence(\n bool ... | 2026-03-21T17:43:35 |
huggingface/transformers | 393b4b3d28e29b4b05b19b4b7f3242a7fc893637 | b92f8ffc6587478b43f33b62f9ff47e4a25cac5c | [Trainer] Move NEFTune impl to standalone functions (#43714)
* style
* update
* Apply suggestion from @kashif
Co-authored-by: Kashif Rasul <kashif.rasul@gmail.com>
* Apply suggestion from @kashif
Co-authored-by: Kashif Rasul <kashif.rasul@gmail.com>
* Apply repo consistency fixes
---------
Co-authored-by: Kash... | [
{
"path": "src/transformers/integrations/__init__.py",
"patch": "@@ -130,6 +130,11 @@\n \"replace_with_mxfp4_linear\",\n \"swizzle_mxfp4\",\n ],\n+ \"neftune\": [\n+ \"activate_neftune\",\n+ \"deactivate_neftune\",\n+ \"neftune_post_forward_hook\",\n+ ],\n ... | 2026-02-05T18:58:11 |
facebook/react | d8089f2cf27a17872a13805fce415b74fd3b107e | 1b4a0daba8aba71f6fbf0c3b88633c51207a6ec7 | [Flight Reply] Encode FormData (#26663)
Builds on top of https://github.com/facebook/react/pull/26661
This lets you pass FormData objects through the Flight Reply
serialization. It does that by prefixing each entry with the ID of the
reference and then the decoding side creates a new FormData object
containing o... | [
{
"path": "fixtures/flight/src/App.js",
"patch": "@@ -7,8 +7,9 @@ import {Counter as Counter2} from './Counter2.js';\n \n import ShowMore from './ShowMore.js';\n import Button from './Button.js';\n+import Form from './Form.js';\n \n-import {like} from './actions.js';\n+import {like, greet} from './actions.j... | 2023-04-18T18:57:33 |
golang/go | 4a3cef2036097d323b6cc0bbe90fc4d8c7588660 | 5254e989425648ff62faa270948bb927b6568ff3 | all: rename crypto/internal/fips to crypto/internal/fips140
Sometimes we've used the 140 suffix (GOFIPS140, crypto/fips140)
and sometimes not (crypto/internal/fips, cmd/go/internal/fips).
Use it always, to avoid having to remember which is which.
Also, there are other FIPS standards, like AES (FIPS 197), SHA-2 (FIPS ... | [
{
"path": "lib/fips140/Makefile",
"patch": "@@ -13,7 +13,7 @@\n #\n # Note that once published a snapshot zip file should never\n # be modified. We record the sha256 hashes of the zip files\n-# in fips140.sum, and the cmd/go/internal/fips test checks\n+# in fips140.sum, and the cmd/go/internal/fips140 test ... | 2024-11-20T14:03:35 |
electron/electron | 841d223b3b04ce247eacaa2fc01a6588ae9b01ae | e119da8ce22eeefc60bdae9cbb040cde8c5f25af | build: rebuild the `dist_zip`s when the deps get modified (#32713)
* build: rebuild the dist_zips when the deps get modified
The dist.zip generated by the electron_dist_zip action was not getting
updated when changes were being made to the dependencies, like the
source files. It turns out, we were using data_deps... | [
{
"path": "BUILD.gn",
"patch": "@@ -1403,11 +1403,13 @@ dist_zip(\"electron_dist_zip\") {\n if (is_linux) {\n data_deps += [ \"//sandbox/linux:chrome_sandbox\" ]\n }\n+ deps = data_deps\n outputs = [ \"$root_build_dir/dist.zip\" ]\n }\n \n dist_zip(\"electron_ffmpeg_zip\") {\n data_deps = [ \"/... | 2022-02-08T18:03:33 |
ggml-org/llama.cpp | 212f4521b013a3eeb79e15df7ca07a5329d39d4b | 568aec82d2fc48341c54cae565768ac75072a31d | context : use n_embd_out for pooled embedding extraction (#20840)
The MEAN/CLS/LAST pooling paths in encode() and decode() used
n_embd_inp() (16384 for qwen3vl with deepstack) to read from the
pooled embedding tensor, which only has n_embd_out() (4096) floats
per sequence. This caused a tensor read out of bounds asser... | [
{
"path": "src/llama-context.cpp",
"patch": "@@ -1347,8 +1347,11 @@ int llama_context::encode(const llama_batch & batch_inp) {\n const llama_seq_id seq_id = ubatch.seq_id_unq[s];\n const int32_t seq_idx = ubatch.seq_idx[seq_id];\n \n- ... | 2026-03-21T17:35:00 |
huggingface/transformers | b92f8ffc6587478b43f33b62f9ff47e4a25cac5c | 759b5c8fd85fb627c40ce8d275041adc0851af2d | [docs] deploying (#43241)
* fix
* feedback
* fix | [
{
"path": "docs/source/en/_toctree.yml",
"patch": "@@ -231,18 +231,24 @@\n - isExpanded: false\n sections:\n - sections:\n+ - local: community_integrations/transformers_as_backend\n+ title: Building a compatible model backend for inference\n - local: community_integrations/sglang\n tit... | 2026-02-05T16:36:48 |
facebook/react | b433c379d55d9684945217c7d375de1082a1abb8 | d962f35cac4e1b9e4c6d9320e60ca51037680296 | Fix input tracking bug (#26627)
In
https://github.com/facebook/react/pull/26573/commits/2019ddc75f448292ffa6429d7625514af192631b,
we changed to set .defaultValue before .value on updates. In some cases,
setting .defaultValue causes .value to change, and since we only set
.value if it has the wrong value, this resu... | [
{
"path": "packages/react-dom-bindings/src/client/ReactDOMComponent.js",
"patch": "@@ -1297,32 +1297,36 @@ export function updateProperties(\n let type = null;\n let value = null;\n let defaultValue = null;\n+ let lastDefaultValue = null;\n let checked = null;\n let defaul... | 2023-04-18T17:49:32 |
golang/go | 5254e989425648ff62faa270948bb927b6568ff3 | 2ad53d5a1928c6afd368c5ff00d2f6c4425e6bcd | cmd/cgo/internal/testsanitizers: fix TSAN tests using setarch
Some systems don't have permissions to run setarch, for example
when running in a docker container without the --privileged flag.
This change makes the tests skip the setarch command if it fails.
Fixes #70463
Cq-Include-Trybots: luci.golang.try:gotip-lin... | [
{
"path": "src/cmd/cgo/internal/testsanitizers/cshared_test.go",
"patch": "@@ -93,12 +93,19 @@ func TestShared(t *testing.T) {\n \n \t\t\tcmdArgs := []string{dstBin}\n \t\t\tif tc.sanitizer == \"thread\" && GOOS == \"linux\" {\n-\t\t\t\t// Disable ASLR for TSAN. See #59418.\n-\t\t\t\tarch, err := exec.Comma... | 2024-11-20T15:02:03 |
electron/electron | 58af7d2a9ab1f3ce7f94b6141a534cb336aa2684 | c09ce25ab6d2315f8dffb5e5c8e10543d695a2f0 | fix: failure to print on macOS (#32767) | [
{
"path": "patches/chromium/.patches",
"patch": "@@ -113,3 +113,4 @@ revert_stop_using_nsrunloop_in_renderer_process.patch\n fix_dont_delete_SerialPortManager_on_main_thread.patch\n feat_add_data_transfer_to_requestsingleinstancelock.patch\n fix_crash_when_saving_edited_pdf_files.patch\n+drop_extra_printing... | 2022-02-08T15:15:30 |
ollama/ollama | 30d7a59ba88b585a172fa86e38df948e839315ee | 85ab552028b8d53fc85561c0c588a80924ecd621 | ollama-debug.c: change 'ld' to 'PRIi64'
* macOS has different definition per info from @mxyng | [
{
"path": "ml/backend/ggml/ggml/src/ollama-debug.c",
"patch": "@@ -1,4 +1,5 @@\n #include <string.h>\n+#include <inttypes.h>\n \n #include \"ollama-debug.h\"\n \n@@ -24,7 +25,7 @@ static void print_tensor(const void *tensor, void (*cb)(const void *, int),\n fprintf(stderr, \"[\");\n for (int i = 0; ... | 2025-03-13T09:10:37 |
ggml-org/llama.cpp | 2bcdddd5e3ade6b1e8c9437a652f9fbcf2ad2512 | eac9c6ea83ff2b72ba3b5459a58c44990823f2cf | fix(rpc): prevent division by zero in deserialize_tensor (#20712)
rpc : prevent division by zero in deserialize_tensor
When receiving an RPC message with a deprecated tensor type (e.g., type 4 or 5 where `blck_size == 0`), `ggml_row_size()` will trigger a division by zero (SIGFPE) and crash the rpc-server.
This pat... | [
{
"path": "ggml/src/ggml-rpc/ggml-rpc.cpp",
"patch": "@@ -1162,12 +1162,18 @@ ggml_tensor * rpc_server::deserialize_tensor(struct ggml_context * ctx, const rp\n return nullptr;\n }\n \n+ // Fix: Prevent division by zero if blck_size is 0 (e.g., deprecated types)\n+ if (ggml_blck_size((enum... | 2026-03-21T13:59:43 |
huggingface/transformers | 759b5c8fd85fb627c40ce8d275041adc0851af2d | e1f376670be2e6a6054d9688376a6097c4541349 | [Trainer] Move sort and rotate checkpoints to standalone functions (#43736)
* move the standalone functions
* update
* style
* update to logger.warning
* Apply repo consistency fixes
---------
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> | [
{
"path": "src/transformers/trainer.py",
"patch": "@@ -24,7 +24,6 @@\n import math\n import os\n import random\n-import re\n import shutil\n import sys\n import tempfile\n@@ -124,8 +123,10 @@\n load_sharded_checkpoint,\n neftune_post_forward_hook,\n number_of_arguments,\n+ rotate_checkpoints,... | 2026-02-05T15:26:44 |
facebook/react | 31f6dd8070aa2f2983bf659e1caf0ce772884d87 | 34bca247d69c60d9318892e335959a3001839a9e | [test] Add failing test showing useMemo bug | [
{
"path": "compiler/forget/src/__tests__/fixtures/compiler/useMemo-inlining-block-return.expect.md",
"patch": "@@ -0,0 +1,46 @@\n+\n+## Input\n+\n+```javascript\n+// @inlineUseMemo\n+function component(a, b) {\n+ let x = useMemo(() => {\n+ if (a) {\n+ return { b };\n+ }\n+ }, [a, b]);\n+ retur... | 2023-04-17T19:59:35 |
golang/go | 2ad53d5a1928c6afd368c5ff00d2f6c4425e6bcd | 91af7119cd33e59a04d96073bc0f40b588938163 | go/types, types2: disallow new methods on (aliases to) cgo-generated types
This makes methods on aliases of cgo-generated types a new compiler error.
That is ok because cgo-behavior is not covered by the G1 compatibility
guarantee.
Background: In 2023 we fixed a gopls issue related to this by actually
enabling method... | [
{
"path": "src/cmd/compile/internal/types2/issues_test.go",
"patch": "@@ -839,7 +839,8 @@ func TestIssue59944(t *testing.T) {\n \ttestenv.MustHaveCGO(t)\n \n \t// The typechecker should resolve methods declared on aliases of cgo types.\n-\tconst src = `\n+\tconst src = `// -gotypesalias=1\n+\n package p\n \... | 2024-11-19T18:28:45 |
ollama/ollama | 85ab552028b8d53fc85561c0c588a80924ecd621 | b3af953a55f0bd054937374404506c4229fbda8c | ollama-debug.c: correct mistype
Signed-off-by: shane.xb.qian <shane.qian@foxmail.com> | [
{
"path": "ml/backend/ggml/ggml/src/ollama-debug.c",
"patch": "@@ -24,7 +24,7 @@ static void print_tensor(const void *tensor, void (*cb)(const void *, int),\n fprintf(stderr, \"[\");\n for (int i = 0; i < dims[0]; i++) {\n if (i >= nitems && i < dims[0] - nitems) {\n- fprintf(stde... | 2025-03-12T14:32:30 |
electron/electron | 81fcd732c2080ca857bc7cb5d5b8c701ab382a90 | d46431b564d9618ca1f057aa8f44f085a0c8550a | fix: savePage throw on relative paths (#32728) | [
{
"path": "docs/api/web-contents.md",
"patch": "@@ -1856,7 +1856,7 @@ the cursor when dragging.\n \n #### `contents.savePage(fullPath, saveType)`\n \n-* `fullPath` string - The full file path.\n+* `fullPath` string - The absolute file path.\n * `saveType` string - Specify the save type.\n * `HTMLOnly` - S... | 2022-02-07T08:51:59 |
huggingface/transformers | e1f376670be2e6a6054d9688376a6097c4541349 | 35b85bd579654f906f110375eb33223885447998 | Fix EP post merge (#43730)
* restore
* add all reduce for ep
* fix init and bias sharding
* fix finalize weight init
* add full stacktracing
* fix
* okay big improvement here
* the only case shard index should be used is when we are acctually collecting for mergeModuleList
* more fixes
* fix EP forward gpt os... | [
{
"path": "docs/source/en/perf_infer_gpu_multi.md",
"patch": "@@ -115,7 +115,7 @@ class ParallelInterface(MutableMapping):\n \"local_colwise\": ColwiseParallel(use_dtensor=False),\n \"local_rowwise\": RowwiseParallel(use_dtensor=False),\n \"local\": IsolatedParallel(),\n- \"ga... | 2026-02-05T15:23:38 |
ggml-org/llama.cpp | 29b28a98242692e352df38254c9a2bb9fa79c884 | cea560f483f0f03e828a6c76e78821debdecbe06 | ci : switch from pyright to ty (#20826)
* type fixes
* switch to ty
* tweak rules
* tweak more rules
* more tweaks
* final tweak
* use common import-not-found rule | [
{
"path": ".github/workflows/python-type-check.yml",
"patch": "@@ -4,35 +4,40 @@ on:\n push:\n paths:\n - '.github/workflows/python-type-check.yml'\n- - 'pyrightconfig.json'\n+ - 'ty.toml'\n - '**.py'\n - '**/requirements*.txt'\n+ # - 'pyrightconfig.json'\n pull_requ... | 2026-03-21T07:54:34 |
facebook/react | 77d3b02e5c8bd92190d20c692fdbf2db01e3657b | b6006201b5fdfcc5720160f169b80ddb7b8d7467 | React DevTools 4.27.4 -> 4.27.5 (#26637)
Full list of changes (not everything included in changelog):
* refactor[devtools]: copy to clipboard only on frontend side
([hoxyq](https://github.com/hoxyq) in
[#26604](https://github.com/facebook/react/pull/26604))
* Provide icon to edge devtools.
([harrygz889](https://g... | [
{
"path": "packages/react-devtools-core/package.json",
"patch": "@@ -1,6 +1,6 @@\n {\n \"name\": \"react-devtools-core\",\n- \"version\": \"4.27.4\",\n+ \"version\": \"4.27.5\",\n \"description\": \"Use react-devtools outside of the browser\",\n \"license\": \"MIT\",\n \"main\": \"./dist/backend.j... | 2023-04-17T16:42:02 |
ollama/ollama | 83f0ec8269eeaaef993af5b61916919db34e8cb7 | c6b6938b3a5d0aef109464e8d42c7849d72bacce | all: address linter errors | [
{
"path": "convert/convert.go",
"patch": "@@ -228,7 +228,6 @@ func ConvertModel(fsys fs.FS, ws io.WriteSeeker) error {\n \tswitch {\n \tcase vocabSize == 0:\n \t\tslog.Warn(\"vocabulary size was not explicitly set by the model\", \"default size\", len(t.Vocabulary.Tokens))\n-\t\tvocabSize = len(t.Vocabulary... | 2025-03-11T18:00:13 |
golang/go | 91af7119cd33e59a04d96073bc0f40b588938163 | d69e6f63c349741a450e9de9585ad555babefad4 | go/types, types2: disallow new methods on generic alias and instantiated types
If the receiver is an alias declaring type parameters, report
an error and ensure that the receiver type remains invalid.
Collect type parameters etc. as before but do not attempt to
find their constraints or instantiate the receiver type.
... | [
{
"path": "src/cmd/compile/internal/types2/resolver.go",
"patch": "@@ -585,19 +585,22 @@ func (check *Checker) unpackRecv(rtyp syntax.Expr, unpackParams bool) (ptr bool,\n \treturn\n }\n \n-// resolveBaseTypeName returns the non-alias base type name for typ, and whether\n+// resolveBaseTypeName returns the ... | 2024-11-18T20:45:17 |
huggingface/transformers | b98123c4966c4821262a312c735c9946edab898b | ffaf03b32acb2876e90fa5c5100e48fecdbf1093 | fix(models): Migrate legacy segmentation_indices to out_indices in BeitConfig (#43505)
* fix: Migrate legacy segmentation_indices to out_indices in BeitConfig
* nit: Fix ci/circleci: check_repository_consistency
* fix: Address reviewer comments | [
{
"path": "src/transformers/models/beit/configuration_beit.py",
"patch": "@@ -151,6 +151,8 @@ def __init__(\n reshape_hidden_states=True,\n **kwargs,\n ):\n+ if \"segmentation_indices\" in kwargs and out_indices is None:\n+ out_indices = kwargs.pop(\"segmentation_indice... | 2026-02-05T13:33:51 |
facebook/react | 2bfe4b246f58d1f8d357f984fba9a8aa1fa79c73 | ab2385fa388f3d4ebc0b1a9f5a747c82e4aa82ed | [Flight] Fix style nit from #26623 (#26629)
Maybe this is faster.
https://github.com/facebook/react/pull/26623#discussion_r1167053174 | [
{
"path": "packages/react-client/src/ReactFlightClient.js",
"patch": "@@ -565,7 +565,7 @@ export function parseModelString(\n }\n case '-': {\n // $-0 or $-Infinity\n- if (value[2] === '0') {\n+ if (value === '$-0') {\n return -0;\n } else {\n re... | 2023-04-14T16:49:41 |
golang/go | d69e6f63c349741a450e9de9585ad555babefad4 | 88cfad0c89014f364e3309e7ee6c4e0d3d382685 | runtime: keep cleanup closure alive across adding the cleanup special
This is similar to the weak handle bug in #70455. In short, there's a
window where a heap-allocated value is only visible through a special
that has not been made visible to the GC yet.
For #70455.
Change-Id: Ic2bb2c60d422a5bc5dab8d971cfc26ff6d762... | [
{
"path": "src/runtime/mheap.go",
"patch": "@@ -2057,6 +2057,11 @@ func addCleanup(p unsafe.Pointer, f *funcval) uint64 {\n \t\t// special isn't part of the GC'd heap.\n \t\tscanblock(uintptr(unsafe.Pointer(&s.fn)), goarch.PtrSize, &oneptrmask[0], gcw, nil)\n \t}\n+\t// Keep f alive. There's a window in thi... | 2024-11-20T19:24:56 |
ggml-org/llama.cpp | b1c70e2e5419ced91eec570b9aabc050afe185e1 | e6ec21e62f45a247d3a3d2cb8f164141853b1a9b | common/parser: fix nasty bug causing subtle corruption of generation prompt (#20825) | [
{
"path": "common/chat-auto-parser-helpers.cpp",
"patch": "@@ -188,6 +188,21 @@ diff_split calculate_diff_split(const std::string & left, const std::string & ri\n result.suffix = \"\";\n // pick prefix = all as representation\n }\n+\n+ // When left has no unique content (result.left i... | 2026-03-20T23:19:04 |
ollama/ollama | c6b6938b3a5d0aef109464e8d42c7849d72bacce | fb4664fcec8c205b48ea7edd0c6f760fc2c346b0 | kvcache: fix tests by adding AvgPool2D stub | [
{
"path": "kvcache/causal_test.go",
"patch": "@@ -445,6 +445,10 @@ func (t *testTensor) AvgPool1D(ctx ml.Context, k, s, p int) ml.Tensor {\n \tpanic(\"not implemented\")\n }\n \n+func (t *testTensor) AvgPool2D(ctx ml.Context, k, s int, p float32) ml.Tensor {\n+\tpanic(\"not implemented\")\n+}\n+\n func (t *... | 2025-03-11T17:57:48 |
huggingface/transformers | a8e791dbc2c4e66f0c921991e3fb42b942c6edc9 | b8a1c69b9681049eebd151446455484b45cc8603 | fix: add compatible_model_types to suppress model type mismatch warnings (#43495)
docs: enhance model type warning messages for compatibility with video variants in EdgeTam, Sam2, and Sam3 configurations | [
{
"path": "src/transformers/configuration_utils.py",
"patch": "@@ -542,8 +542,11 @@ def from_pretrained(\n # raise warning only if we still can't see a match in `model_type`\n if config_dict[\"model_type\"] != cls.model_type:\n logger.warning(\n- f\... | 2026-02-05T13:03:32 |
golang/go | 88cfad0c89014f364e3309e7ee6c4e0d3d382685 | e0b569842d0ef5deed5c455eef001b02aec27934 | runtime: explicitly keep handle alive during getOrAddWeakHandle
getOrAddWeakHandle is very careful about keeping its input alive across
the operation, but not very careful about keeping the heap-allocated
handle it creates alive. In fact, there's a window in this function
where it is *only* visible via the special. Sp... | [
{
"path": "src/runtime/mheap.go",
"patch": "@@ -2224,8 +2224,14 @@ func getOrAddWeakHandle(p unsafe.Pointer) *atomic.Uintptr {\n \n \t\t// Keep p alive for the duration of the function to ensure\n \t\t// that it cannot die while we're trying to do this.\n+\t\t//\n+\t\t// Same for handle, which is only store... | 2024-11-20T19:12:58 |
ggml-org/llama.cpp | 149b2493c022c621d16154b6d261769da35e9e00 | b31b30f31db73018cee86ccd20c9ad132452e6a5 | common : fix typo in debug log ('extracft' -> 'extract') (#20807) | [
{
"path": "common/chat-diff-analyzer.cpp",
"patch": "@@ -409,7 +409,7 @@ void analyze_reasoning::compare_reasoning_scope() {\n if (result.result.success()) {\n end = trim_trailing_whitespace(result.tags[\"post\"]);\n } else {\n- LOG_DBG(ANSI_ORANGE \"%s... | 2026-03-20T17:23:18 |
ollama/ollama | 06007c0a187f3fa12cde58fc337f15cbf7a6cbf8 | a8e83a7654fffa169b90fa927e6d19c4c0c765d7 | Allow models to force a new batch
This is useful for a few things:
- Work around bugs, such as having 2 images in one batch
- Keep the image in a single batch for fully connected attention
- Improve performance by not evaluating embeddings multiple times | [
{
"path": "model/input/input.go",
"patch": "@@ -15,6 +15,12 @@ type Input struct {\n \t// stored in Multimodal, used for caching and comparing\n \t// equality.\n \tMultimodalHash uint64\n+\n+\t// BatchBreak forces a new batch to be started with this\n+\t// input. For example, this can be used to align image... | 2025-03-11T03:03:29 |
electron/electron | c3d11e2ea2d9751a457c72805b7688a38ceac30a | ed185f324e791da5dea3475aeb8d4fae798d7122 | test: fix failing tests of focus/blur events of WebContents (#32711) | [
{
"path": "docs/api/web-contents.md",
"patch": "@@ -516,6 +516,15 @@ Emitted when the `WebContents` loses focus.\n \n Emitted when the `WebContents` gains focus.\n \n+Note that on macOS, having focus means the `WebContents` is the first responder\n+of window, so switching focus between windows would not tri... | 2022-02-02T15:32:53 |
huggingface/transformers | b8a1c69b9681049eebd151446455484b45cc8603 | ace7c370145a4a4861383672daa7095a0c8e43fc | Fix T5 v1.1 detection (#43681)
* Fix T5 v1.1 detection
PR #41541 refactored `tie_word_embeddings` handling (among other things)
which subtly broke detection of T5 v1.1 vs. original detection. As a
consequence, decoder output scaling was always applied, regardless of
T5 version.
This is resolved by using the correct ... | [
{
"path": "src/transformers/models/t5/configuration_t5.py",
"patch": "@@ -140,7 +140,7 @@ def __init__(\n # The model code was relying on saved configs where `tie_word_embeddings` is\n # set to `False` in 1.1v and using it as indicator of whether to scale or not\n # But in fact we ti... | 2026-02-05T11:02:02 |
golang/go | e36b4d76ad697536921f2bcd119e7cf2a702b920 | dc28aca56830f8eca7005d045cbcc438cc1e8fe3 | doc/godebug: document we removed x509sha1
Change-Id: I4a4ff0b9e0d3584b96e414d60b889d68fa8660c0
Reviewed-on: https://go-review.googlesource.com/c/go/+/630236
Auto-Submit: Roland Shoemaker <roland@golang.org>
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-acco... | [
{
"path": "doc/godebug.md",
"patch": "@@ -383,7 +383,7 @@ There is no plan to remove this setting.\n \n Go 1.18 removed support for SHA1 in most X.509 certificates,\n controlled by the [`x509sha1` setting](/pkg/crypto/x509#InsecureAlgorithmError).\n-This setting will be removed in a future release, Go 1.22 ... | 2024-11-20T19:06:35 |
ggml-org/llama.cpp | 58c81f7e811d82b9ba62002f40d7af56134b61fc | fb78ad29bbe7ae00619b2ce31b0a71e95fdbfc43 | model : fix Granite Hybrid type check for 7B.A1B (#20795)
* Check granite hybriid expert count to set type as LLM_TYPE_7B_A1B or LLM_TYPE_1B
* Use feed fwd dim instead of num of experts
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
---------
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scal... | [
{
"path": "src/llama-model.cpp",
"patch": "@@ -2058,7 +2058,7 @@ void llama_model::load_hparams(llama_model_loader & ml) {\n \n switch (hparams.n_embd) {\n case 768: type = LLM_TYPE_350M; break;\n- case 1536: type = (hparams.n_embd == 2048 ? LLM_TYPE_7B... | 2026-03-20T14:16:09 |
ollama/ollama | 2c40c4d35eddc86673c5b0c116e2a34ef8ee2c4a | e95278932bfd0eac5800ec28efe5c49a0d9ce8c3 | Fix follow up images and images split across batches | [
{
"path": "model/models/gemma3/model.go",
"patch": "@@ -5,7 +5,6 @@ import (\n \t\"encoding/binary\"\n \t\"hash/fnv\"\n \t\"image\"\n-\t\"slices\"\n \n \t\"github.com/ollama/ollama/kvcache\"\n \t\"github.com/ollama/ollama/ml\"\n@@ -99,49 +98,43 @@ func (m *Model) EncodeMultimodal(ctx ml.Context, multimodalD... | 2025-03-10T04:29:58 |
vuejs/vue | 9e88707940088cb1f4cd7dd210c9168a50dc347c | 73486cb5f5862a443b42c2aff68b82320218cbcd | chore: fix spelling mistakes (#13216)
[skip ci] | [
{
"path": "CHANGELOG.md",
"patch": "@@ -3,7 +3,7 @@\n \n ### Bug Fixes\n \n-* **lifecycle:** esnure component effect scopes are disconnected ([56ce7f8](https://github.com/vuejs/vue/commit/56ce7f8c573116ed6683149206cf35c987249d42)), closes [#13134](https://github.com/vuejs/vue/issues/13134)\n+* **lifecycle:*... | 2024-10-10T07:24:14 |
electron/electron | 9a5a45e804d83ea06f458d2ff9020aec87ffc17d | 8b6202b6a8847ddd28768fe09b0e014259a6b8e7 | fix: WCO window hover on window controls on Windows (#32672)
* fix: WCO window hover on window controls
* Update shell/browser/ui/win/electron_desktop_window_tree_host_win.cc
Co-authored-by: Robo <hop2deep@gmail.com>
Co-authored-by: Robo <hop2deep@gmail.com> | [
{
"path": "shell/browser/ui/win/electron_desktop_window_tree_host_win.cc",
"patch": "@@ -99,4 +99,21 @@ bool ElectronDesktopWindowTreeHostWin::GetClientAreaInsets(\n return false;\n }\n \n+bool ElectronDesktopWindowTreeHostWin::HandleMouseEvent(ui::MouseEvent* event) {\n+ // Call the default implementati... | 2022-02-02T15:06:36 |
huggingface/transformers | ace7c370145a4a4861383672daa7095a0c8e43fc | 83bce8d84016efda1e1c838e9da34c302c718701 | Add moonshine streaming (#43702)
* Add Moonshine Streaming
* sliding window attention in encoder (first 2 and last 2 layers)
* different dims for encoder and decoder
* projection and position embedding addition in adapter
* Move fully defined MoonshineStreamingConfig to configuration
Not auto generated anymore.
* ... | [
{
"path": "docs/source/en/_toctree.yml",
"patch": "@@ -963,6 +963,8 @@\n title: MMS\n - local: model_doc/moonshine\n title: Moonshine\n+ - local: model_doc/moonshine_streaming\n+ title: Moonshine Streaming\n - local: model_doc/moshi\n title: Moshi\n - lo... | 2026-02-04T18:29:51 |
golang/go | dc28aca56830f8eca7005d045cbcc438cc1e8fe3 | e1dc707f33b0cecae32e6e649fbd58a7274fc322 | crypto/ed25519: fix GenerateKey with rand nil
GenerateKey(nil) is documented to use crypto/rand.Reader, but we didn't
have a test.
While at it, since it's documented to be equivalent to NewKeyFromSeed,
actually implement it that way. This has the probably good side effect
of making it deterministic in FIPS mode. The ... | [
{
"path": "src/crypto/ed25519/ed25519.go",
"patch": "@@ -18,6 +18,7 @@ package ed25519\n import (\n \t\"crypto\"\n \t\"crypto/internal/fips/ed25519\"\n+\tcryptorand \"crypto/rand\"\n \t\"crypto/subtle\"\n \t\"errors\"\n \t\"io\"\n@@ -130,17 +131,17 @@ func (o *Options) HashFunc() crypto.Hash { return o.Hash... | 2024-11-20T18:25:37 |
ggml-org/llama.cpp | e06c3ab2bc7e45df7584468014681349fceccfbc | dc6592431b909208040c1a8e953e6c5440471eaa | vulkan: change gated_delta_net to shard a column across a subgroup (#20662)
* vulkan: change gated_delta_net to shard a column across a subgroup
This is based on https://github.com/ggml-org/llama.cpp/pull/20391, I used an
LLM to port the CUDA code to Vulkan, and guided to it to make various fixes to
work with Vulkan ... | [
{
"path": "ggml/src/ggml-vulkan/ggml-vulkan.cpp",
"patch": "@@ -4604,12 +4604,42 @@ static void ggml_vk_load_shaders(vk_device& device) {\n {\"gated_delta_net_f32_d64\", \"gated_delta_net_f32_d64_kda\"},\n {\"gated_delta_net_f32_d128\", \"gated_delta_net_f32_d128_kda\"},\n ... | 2026-03-20T11:17:15 |
ollama/ollama | 2e54d72fc36f85d45c4392f14b2433972736d035 | 6b32a2d549ca3c37cae64ae6d6d2419e0f2bc139 | fix gemma3 1b conversion | [
{
"path": "convert/convert_gemma3.go",
"patch": "@@ -80,12 +80,13 @@ func (p *gemma3Model) KV(t *Tokenizer) ggml.KV {\n \t\tkv[\"gemma3.attention.key_length\"] = p.HeadDim\n \t\tkv[\"gemma3.attention.value_length\"] = p.HeadDim\n \t\tkv[\"gemma3.attention.sliding_window\"] = p.SlidingWindow\n-\t\tkv[\"gemma... | 2025-03-10T22:35:07 |
huggingface/transformers | 83bce8d84016efda1e1c838e9da34c302c718701 | 452c179eb631aeb4a08b4dba647c2bc727537a1d | Allow bi-directional attention for all models (#43705)
* mask
* decorator
* oupsi
* fix
* fix
* use None as default
* add test
* add fa2
* device | [
{
"path": "src/transformers/masking_utils.py",
"patch": "@@ -865,6 +865,17 @@ def create_causal_mask(\n An optional mask function to combine with the causal mask function (by doing the intersection of both). This is\n useful to easily overlay another mask on top of the causal one, fo... | 2026-02-04T17:24:28 |
vuejs/vue | 73486cb5f5862a443b42c2aff68b82320218cbcd | e428d891456eeb6d34cfac070c423694dcda8806 | chore: fix link broken (#13192)
* chore: fix link broken
Signed-off-by: snoppy <michaleli@foxmail.com>
* Update packages/template-compiler/README.md [skip ci]
---------
Signed-off-by: snoppy <michaleli@foxmail.com>
Co-authored-by: Eduardo San Martin Morote <posva@users.noreply.github.com> | [
{
"path": "packages/template-compiler/README.md",
"patch": "@@ -130,7 +130,7 @@ This is only useful at runtime with pre-configured builds, so it doesn't accept\n \n Same as `compiler.compile` but generates SSR-specific render function code by optimizing parts of the template into string concatenation in ord... | 2024-06-14T12:52:12 |
electron/electron | 7c701367c0820691e2a0f8f1cfd1fb3470071704 | 56c6d25e98d6c0caef124dc81113f264efccc8ef | fix: some frameless windows showing a frame on Windows (#32692) | [
{
"path": "shell/browser/native_window_views.cc",
"patch": "@@ -329,15 +329,14 @@ NativeWindowViews::NativeWindowViews(const gin_helper::Dictionary& options,\n // Set Window style so that we get a minimize and maximize animation when\n // frameless.\n DWORD frame_style = WS_CAPTION | WS_OVERLAPP... | 2022-02-02T10:23:02 |
ggml-org/llama.cpp | ab9d4c3678a6f8bb797610a27bc0af493fcf786c | 1af9dab32b555f7e3aac8b1bfe4e4cb780cb4185 | server : improve mtmd ctx checkpoints (#20726)
* server : improve mtmd ctx checkpoints
* server : fix off-by-one in pos_min_thold | [
{
"path": "tools/server/server-context.cpp",
"patch": "@@ -2307,8 +2307,8 @@ struct server_context_impl {\n \n llama_pos pos_next = slot.prompt.tokens.pos_next(n_past);\n \n- // note: when n_swa == 0, the model does not use SWA, which is equivalent to a... | 2026-03-20T09:13:12 |
golang/go | e1dc707f33b0cecae32e6e649fbd58a7274fc322 | 7c7170e93934d78b1a1c5c0288f5cdbbf812eeee | cmd/go/testdata/script: restrict test to check only exe name
Avoid test failures caused by 'v2' in user environment paths.
Modify the test to check only the output executable name and ensure it is not 'v2', rather than inspecting the entire path.
Fixes #67989
Change-Id: Ida2131f6c9b41724df1b6b5e31413252c5009d25
Revi... | [
{
"path": "src/cmd/go/testdata/script/mod_install_versioned.txt",
"patch": "@@ -7,7 +7,7 @@ stdout 'fortune(\\.exe)?$'\n \n go get rsc.io/fortune/v2\n go list -f '{{.Target}}' rsc.io/fortune/v2\n-! stdout v2\n+! stdout 'v2(\\.exe)?$'\n stdout 'fortune(\\.exe)?$'\n \n -- go.mod --",
"additions": 1,
"... | 2024-10-21T04:58:08 |
ollama/ollama | f888912870c16bf11511880775132804c441e731 | 9e4642e9b3e9a26d423c62915805375ca253d7d1 | fix vision encoder | [
{
"path": "model/models/gemma3/model_text.go",
"patch": "@@ -180,7 +180,7 @@ func (m *TextModel) Forward(ctx ml.Context, inputs, positions, outputs ml.Tensor\n \tif multimodal != nil {\n \t\tvisionOutputs := multimodal[0].Multimodal.(ml.Tensor)\n \t\toffset := multimodal[0].Index - 1 - visionOutputs.Dim(1)\... | 2025-03-09T23:18:13 |
ggml-org/llama.cpp | 6d99b44c7eb868e8b9d49d0b4cab243dd70a4b53 | 464fd0e71f13b9ed7676965faa35f7aee02d8c86 | docs : fix Metal backend op support status in ops.md (#20779)
Regenerate docs/ops/Metal.csv using test-backend-ops on Apple M5
and rebuild docs/ops.md via scripts/create_ops_docs.py.
Five ops were incorrectly marked as not supported (❌) for Metal:
- DIAG: ❌ → ✅
- POOL_1D: ❌ → ✅
- SET: ❌ → ... | [
{
"path": "docs/ops.md",
"patch": "@@ -12,9 +12,9 @@ Legend:\n - 🟡 Partially supported by this backend\n - ❌ Not supported by this backend\n \n-| Operation | BLAS | CANN | CPU | CUDA | Metal | OpenCL | SYCL | Vulkan | WebGPU | ZenDNN | zDNN |\n+| Operation | BLAS | CANN | CPU | CUDA | MTL | OpenCL | SYCL |... | 2026-03-20T09:06:38 |
huggingface/transformers | 452c179eb631aeb4a08b4dba647c2bc727537a1d | 1744f8f3fc19dfb812887afbe5651eb49995bed5 | Docs: fix Training step by removing tokenizer from trainer initialization (#43733)
* Docs: fix Training step by removing tokenizer from trainer initialization
We already provide the tokenizer in the data_collator. Trying to provide the tokenizer parameter in the Trainer() __init__() method produces the following erro... | [
{
"path": "docs/source/en/quicktour.md",
"patch": "@@ -252,7 +252,7 @@ trainer = Trainer(\n args=training_args,\n train_dataset=dataset[\"train\"],\n eval_dataset=dataset[\"test\"],\n- tokenizer=tokenizer,\n+ processing_class=tokenizer,\n data_collator=data_collator,\n )\n ",
"addi... | 2026-02-04T16:43:58 |
vuejs/vue | 56ce7f8c573116ed6683149206cf35c987249d42 | 305e4ae9dce0a05a422b675ff555f17878e86063 | fix(lifecycle): esnure component effect scopes are disconnected
close #13134 | [
{
"path": "src/core/instance/init.ts",
"patch": "@@ -34,6 +34,9 @@ export function initMixin(Vue: typeof Component) {\n vm.__v_skip = true\n // effect scope\n vm._scope = new EffectScope(true /* detached */)\n+ // #13134 edge case where a child component is manually created during the\n+ /... | 2023-12-24T14:40:05 |
electron/electron | 56c6d25e98d6c0caef124dc81113f264efccc8ef | b346f909e74dc3a82df2ba269029dc3dbd365add | fix: crash on printer dialog cancellation (#32632)
* fix: crash on printer dialog cancellation
* fix: remove commented out code
* chore: address review | [
{
"path": "patches/chromium/printing.patch",
"patch": "@@ -11,7 +11,7 @@ majority of changes originally come from these PRs:\n This patch also fixes callback for manual user cancellation and success.\n \n diff --git a/chrome/browser/printing/print_job.cc b/chrome/browser/printing/print_job.cc\n-index 0de053... | 2022-02-01T19:00:09 |
ollama/ollama | 9e4642e9b3e9a26d423c62915805375ca253d7d1 | 6b0486c216d639257b48f7c0b0bb01358bdbc765 | ollama debug tensor | [
{
"path": "llama/patches/0020-ollama-debug-tensor.patch",
"patch": "@@ -0,0 +1,33 @@\n+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001\n+From: Michael Yang <mxyng@pm.me>\n+Date: Sun, 9 Mar 2025 14:44:16 -0700\n+Subject: [PATCH] ollama debug tensor\n+\n+---\n+ ggml/src/ggml-cpu/ggml-cp... | 2025-03-09T21:45:36 |
golang/go | 7c7170e93934d78b1a1c5c0288f5cdbbf812eeee | c901d93fcd5b1a2439232de4469400fd7a327128 | crypto/x509: remove x509sha1 GODEBUG
Fixes #41682
Change-Id: I37760f2186e75ec7df9674db25ae466cf453d66d
Reviewed-on: https://go-review.googlesource.com/c/go/+/629676
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Filippo Valsorda <filippo@golang.org>
Reviewed-by:... | [
{
"path": "doc/godebug.md",
"patch": "@@ -188,6 +188,9 @@ calling into Go code from C will enable DIT, and disable it before returning to\n C if it was not enabled when Go code was entered.\n This currently only affects arm64 programs. For all other platforms it is a no-op.\n \n+Go 1.24 removed the `x509sha... | 2024-11-19T19:34:19 |
huggingface/transformers | 1744f8f3fc19dfb812887afbe5651eb49995bed5 | 8dce31003b16946d0e2ee035b94a5e73e7dee7cd | Fix scheduler initialization order (#43711)
* Fix scheduler
* style
* fix
* fix | [
{
"path": "src/transformers/integrations/deepspeed.py",
"patch": "@@ -549,8 +549,6 @@ def _lr_scheduler_callable(optimizer):\n return lr_scheduler\n \n lr_scheduler = DummyScheduler(optimizer, lr_scheduler_callable=_lr_scheduler_callable)\n- else:\n- lr_schedule... | 2026-02-04T16:08:17 |
ggml-org/llama.cpp | 21c804521479c750bdf6d26506964ab3aa207481 | c46583b86bed573c4ff30685dae59874f124e664 | jinja : fix heap OOB read in value equality comparison (#20782)
Address GHSA-q9j6-4hhc-rq9p and GHSA-2q4c-9gq5-5vfp.
The three-iterator overload of std::equal in value_array_t::equivalent()
and value_object_t::equivalent() reads past the end of the shorter
container when comparing arrays or objects of different lengt... | [
{
"path": "common/jinja/value.h",
"patch": "@@ -451,7 +451,7 @@ struct value_array_t : public value_t {\n }\n protected:\n virtual bool equivalent(const value_t & other) const override {\n- return typeid(*this) == typeid(other) && is_hashable() && other.is_hashable() && std::equal(val_arr.beg... | 2026-03-20T06:15:17 |
electron/electron | b346f909e74dc3a82df2ba269029dc3dbd365add | 939bfa50f6c9b8a497b7147b3412081805fa0af4 | ci: fixup gclient cache on ubuntu 20 (#32691)
* ci: fixup gclient cache on ubuntu 20
* ci: update to Ubuntu 20.04 docker image | [
{
"path": ".circleci/build_config.yml",
"patch": "@@ -45,7 +45,7 @@ executors:\n type: enum\n enum: [\"medium\", \"xlarge\", \"2xlarge+\"]\n docker:\n- - image: ghcr.io/electron/build:27db4a3e3512bfd2e47f58cea69922da0835f1d9\n+ - image: ghcr.io/electron/build:e6bebd08a51a0d78ec... | 2022-02-01T16:54:27 |
vuejs/vue | db9c566032da0ec5cd758a3e8525e9733874c1e5 | 895669ffa01f8075a326308caa8ad6a5f69a1919 | fix: account for nested render calls
close #13131 | [
{
"path": "src/core/instance/render.ts",
"patch": "@@ -15,7 +15,7 @@ import VNode, { createEmptyVNode } from '../vdom/vnode'\n \n import { isUpdatingChildComponent } from './lifecycle'\n import type { Component } from 'types/component'\n-import { setCurrentInstance } from 'v3/currentInstance'\n+import { cur... | 2023-12-14T01:21:11 |
ollama/ollama | 9b54267e6951ec6cff7fc2188899ac5d4169ab79 | 46bb0169c40f466994cde8a89b455d8e3f9e177e | fix configs | [
{
"path": "convert/convert_gemma3.go",
"patch": "@@ -10,6 +10,7 @@ type gemma3Model struct {\n \tgemmaModel\n \tArchitecture string\n \tTextModel struct {\n+\t\tHeadDim uint32 `json:\"head_dim\"`\n \t\tHiddenSize uint32 `json:\"hidden_size\"`\n \t\tHiddenLayers uint32 `json:\"num_hidde... | 2025-03-09T00:16:23 |
golang/go | c901d93fcd5b1a2439232de4469400fd7a327128 | 30b09bccfd2446de566f9dfd0592cb4406e639b2 | cmd/go: enforce -Wl,--push-state logic only when it is a prefix
Make sure we only run the --push-state logic when -Wl,--push-state is a
prefix of the argument, not just present in the argument string.
Thanks to Juho Forsén of Mattermost for reporting this issue.
Change-Id: I799f7854ff680674fd84cf2136fadf70817fc7e2
R... | [
{
"path": "src/cmd/go/internal/work/security.go",
"patch": "@@ -312,7 +312,7 @@ Args:\n \t\tfor _, re := range valid {\n \t\t\tif match := re.FindString(arg); match == arg { // must be complete match\n \t\t\t\tcontinue Args\n-\t\t\t} else if match == \"-Wl,--push-state\" {\n+\t\t\t} else if strings.HasPrefi... | 2024-11-15T15:40:49 |
huggingface/transformers | 8dce31003b16946d0e2ee035b94a5e73e7dee7cd | d75266f17e89b1618ac93802b0c4ced0f2a014bc | Fix accelerate integration import (#43732)
* move function to transformers
* fix | [
{
"path": "src/transformers/integrations/accelerate.py",
"patch": "@@ -44,7 +44,7 @@\n if is_accelerate_available():\n from accelerate import dispatch_model\n from accelerate.utils import get_max_memory\n- from accelerate.utils.modeling import clean_device_map, get_max_layer_size, get_module_size... | 2026-02-04T14:57:43 |
ggml-org/llama.cpp | c46583b86bed573c4ff30685dae59874f124e664 | c1b911654a581662c19eb51829705e4ed7020673 | common/parser : fix out_of_range crash in throw path (#20424 regression) (#20777)
* chat : fix out_of_range crash in throw path (#20424 regression)
#20424 introduced effective_input = generation_prompt + input, but the
throw path uses input.substr(result.end) where result.end is a position
within effective_input. Eve... | [
{
"path": "common/chat.cpp",
"patch": "@@ -1774,7 +1774,7 @@ common_chat_msg common_chat_peg_parse(const common_peg_arena & src_pars\n return msg;\n }\n throw std::runtime_error(std::string(\"Failed to parse input at pos \") + std::to_string(result.end) + \": \" +\n- ... | 2026-03-20T01:37:22 |
electron/electron | 2289a52fb30c2eacdf66333cd163774d00819c75 | aeee9cfb7876a770fed9a1dc9669c2beeacf9ed9 | docs: fix typo in BrowserWindow description (#32667) | [
{
"path": "docs/api/browser-window.md",
"patch": "@@ -388,7 +388,7 @@ It creates a new `BrowserWindow` with native properties as set by the `options`.\n contain the layout of the document—without requiring scrolling. Enabling\n this will cause the `preferred-size-changed` event to be emitted on ... | 2022-02-01T11:47:49 |
vuejs/vue | 895669ffa01f8075a326308caa8ad6a5f69a1919 | 73bdf14ef5c9979dcba18ede3410515cecbe2f2f | fix(types): export more types for v3 alignment (jsx / component options)
close #13078
close #13128 | [
{
"path": "types/index.d.ts",
"patch": "@@ -1,6 +1,7 @@\n import { Vue } from './vue'\n import './umd'\n import './jsx'\n+export * from './jsx'\n \n export default Vue\n \n@@ -49,7 +50,12 @@ export {\n ComputedOptions as ComponentComputedOptions,\n MethodOptions as ComponentMethodOptions,\n ComponentP... | 2023-12-14T00:33:45 |
ollama/ollama | 0e886595bf3d4ee33737f4b30154210b0df2d2df | c62861f4fa4bd5a82a1c8b46bea74b1672d384b3 | Fix tests and drift from main | [
{
"path": "kvcache/causal_test.go",
"patch": "@@ -499,6 +499,10 @@ func (t *testTensor) Contiguous(ctx ml.Context) ml.Tensor {\n \tpanic(\"not implemented\")\n }\n \n+func (t *testTensor) Set(ctx ml.Context, t2 ml.Tensor, offset int, strides ...int) ml.Tensor {\n+\tpanic(\"not implemented\")\n+}\n+\n func (... | 2025-03-08T00:05:56 |
golang/go | 30b09bccfd2446de566f9dfd0592cb4406e639b2 | 5165f54167a404985428cfd87aecc80595af491a | crypto/internal/fipstest: fix TestACVP not to write to module cache
Not sure how this ever worked; the module cache is read-only.
Change-Id: I22e4945fa0bac101995ab9101d22dcfdcee40776
Reviewed-on: https://go-review.googlesource.com/c/go/+/629976
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Reviewed-by: Filippo ... | [
{
"path": "src/crypto/internal/fipstest/acvp_test.go",
"patch": "@@ -368,9 +368,11 @@ func TestACVP(t *testing.T) {\n \tt.Log(\"building acvptool\")\n \n \t// Build the acvptool binary.\n+\ttoolPath := filepath.Join(t.TempDir(), \"acvptool.exe\")\n \tgoTool := testenv.GoToolPath(t)\n \tcmd := testenv.Comman... | 2024-11-20T15:08:18 |
huggingface/transformers | d75266f17e89b1618ac93802b0c4ced0f2a014bc | a276a763fa785a7f429c5ce8241f5b16064b2f44 | 🚨 T5Gemma2 model structure (#43633)
* maybe
* fix repo
* fix test
* do conversion but can't tests, hub is down?
* oops
* ah the model was gated!
* override helper
* fix repo and test
* Update src/transformers/models/t5gemma2/modular_t5gemma2.py
Co-authored-by: Anton Vlasjuk <73884904+vasqu@users.noreply.gith... | [
{
"path": "src/transformers/conversion_mapping.py",
"patch": "@@ -70,6 +70,16 @@\n \n def _build_checkpoint_conversion_mapping():\n mapping = {\n+ \"t5gemma2\": [\n+ WeightRenaming(r\"(?<!vision_model\\.)encoder.embed_tokens.\", \"encoder.text_model.embed_tokens.\"),\n+ Weig... | 2026-02-04T14:44:53 |
ggml-org/llama.cpp | c1b911654a581662c19eb51829705e4ed7020673 | b739738dadf0b66a59546d7240c554d61c07c2f0 | server: fix router mode deadlock on child crash and TOCTOU race in models_max (#20763)
Two bugs in `server_models::load()` that affect router mode reliability:
**Bug 1: Deadlock when child process crashes**
When a child process is killed (e.g., SIGKILL from OS code signature
validation), the monitoring thread deadlo... | [
{
"path": "tools/server/server-models.cpp",
"patch": "@@ -539,6 +539,22 @@ void server_models::load(const std::string & name) {\n return;\n }\n \n+ // Re-check capacity under the lock to prevent concurrent loads from\n+ // exceeding models_max. Without this, the window between unload_lru()... | 2026-03-19T21:16:05 |
vuejs/vue | e0747f40a879b4000a1959d21377b51d1f1ed988 | 2632249925e632e56f6dfc8fdbcf682c82e4081b | fix(keep-alive): fix memory leak without breaking transition tests | [
{
"path": "src/core/components/keep-alive.ts",
"patch": "@@ -33,10 +33,15 @@ function matches(\n }\n \n function pruneCache(\n- keepAliveInstance: { cache: CacheEntryMap; keys: string[]; _vnode: VNode },\n+ keepAliveInstance: {\n+ cache: CacheEntryMap\n+ keys: string[]\n+ _vnode: VNode\n+ $vno... | 2023-12-07T09:07:15 |
ollama/ollama | c62861f4fa4bd5a82a1c8b46bea74b1672d384b3 | 0df1800436bfb48b0556fce8f37c5d2e256c1d0a | fix conversion | [
{
"path": "convert/convert.go",
"patch": "@@ -190,8 +190,8 @@ func ConvertModel(fsys fs.FS, ws io.WriteSeeker) error {\n \t\tconv = &gemmaModel{}\n \tcase \"Gemma2ForCausalLM\":\n \t\tconv = &gemma2Model{}\n-\tcase \"Gemma3ForConditionalGeneration\":\n-\t\tconv = &gemma3Model{}\n+\tcase \"Gemma3ForCausalLM\... | 2025-03-07T22:06:10 |
electron/electron | aeee9cfb7876a770fed9a1dc9669c2beeacf9ed9 | e34d7f5d6f7c34332a11010ca786e2e1271b2629 | feat: add focus and blur events for WebContents (#25873)
test: add focus and blur WebContents event tests
test: confirm that webcontents focus event is fired on browserwindow focus
fix: mac focus event test timeout | [
{
"path": "docs/api/web-contents.md",
"patch": "@@ -508,6 +508,14 @@ Returns:\n \n Emitted when the user is requesting to change the zoom level using the mouse wheel.\n \n+#### Event: 'blur'\n+\n+Emitted when the `WebContents` loses focus.\n+\n+#### Event: 'focus'\n+\n+Emitted when the `WebContents` gains f... | 2022-02-01T10:28:57 |
golang/go | 297081eb021d3c66854f21a93462553e518988cc | c483fdbfcfba9c86dbe93a096f120cc15f783d38 | crypto/internal/fips/rsa: support all SHA hashes in PKCS#1 v1.5
The byte sequences match those in
https://github.com/randombit/botan/blob/e5ec40828/src/lib/pk_pad/hash_id/hash_id.cpp
For #69536
Fixes #43923
Change-Id: I8b4daea71c2f696ad67ddc13affefd1563c51266
Reviewed-on: https://go-review.googlesource.com/c/go/+/63... | [
{
"path": "src/crypto/internal/fips/rsa/pkcs1v15.go",
"patch": "@@ -23,14 +23,20 @@ import (\n // precompute a prefix of the digest value that makes a valid ASN1 DER string\n // with the correct contents.\n var hashPrefixes = map[string][]byte{\n-\t\"MD5\": {0x30, 0x20, 0x30, 0x0c, 0x06, 0x08, 0x2a, ... | 2024-11-20T14:02:40 |
ggml-org/llama.cpp | 340807273b6aa765c9353804b7ce680920373cb6 | 26c9ce128825ba53a91baf75b5b817a1373b46bf | hip: Avoid compiler bug in RDNA code generation during debug builds on Windows (#20655) | [
{
"path": "ggml/src/ggml-hip/CMakeLists.txt",
"patch": "@@ -129,6 +129,11 @@ endif()\n \n if (CXX_IS_HIPCC)\n set_source_files_properties(${GGML_SOURCES_ROCM} PROPERTIES LANGUAGE CXX)\n+ if (WIN32 AND CMAKE_BUILD_TYPE STREQUAL \"Debug\")\n+ # CMake on Windows doesn't support the HIP language y... | 2026-03-19T18:14:08 |
huggingface/transformers | a276a763fa785a7f429c5ce8241f5b16064b2f44 | d95672f8dfe5a5f3f2d4a7a1476d51b33ea835a6 | Update torch minimum version to 2.4 (#41307)
* Update torch minimum version to 2.4
Signed-off-by: Yuanyuan Chen <cyyever@outlook.com>
* More PyTorch 2.4 fixes
Signed-off-by: cyy <cyyever@outlook.com>
* More PyTorch 2.4 fixes
Signed-off-by: cyy <cyyever@outlook.com>
* More fixes
Signed-off-by: cyy <cyyever@outlo... | [
{
"path": "README.md",
"patch": "@@ -81,7 +81,7 @@ Explore the [Hub](https://huggingface.com/) today to find a model and use Transf\n \n ## Installation\n \n-Transformers works with Python 3.9+, and [PyTorch](https://pytorch.org/get-started/locally/) 2.1+.\n+Transformers works with Python 3.9+, and [PyTorch... | 2026-02-04T14:12:30 |
vuejs/vue | 2632249925e632e56f6dfc8fdbcf682c82e4081b | 3650c12f7d3a20f3155bc1fd2b068e84289e0d33 | fix(keep-alive): fix keep-alive memory leak
close #12827 | [
{
"path": "src/core/vdom/create-component.ts",
"patch": "@@ -62,6 +62,8 @@ const componentVNodeHooks = {\n vnode, // new parent vnode\n options.children // new children\n )\n+ // #12187 unset children reference after use to avoid memory leak\n+ options.children = undefined\n },\n \n ... | 2023-12-07T08:12:36 |
electron/electron | e34d7f5d6f7c34332a11010ca786e2e1271b2629 | f5dc2a653528fb84ad8aac1fb4f2a7075f35ad2f | docs: fix missing tag end in clipboard example (#32673) | [
{
"path": "docs/api/clipboard.md",
"patch": "@@ -76,7 +76,7 @@ Writes `markup` to the clipboard.\n ```js\n const { clipboard } = require('electron')\n \n-clipboard.writeHTML('<b>Hi</b')\n+clipboard.writeHTML('<b>Hi</b>')\n ```\n \n ### `clipboard.readImage([type])`",
"additions": 1,
"deletions": 1,
... | 2022-02-01T08:57:33 |
ollama/ollama | 4346c2409dec7bb182aa91d236f0759d592a1b42 | 4b037a97dcc0387d2b94d293d4bd20e7e262d609 | fix drift from main | [
{
"path": "kvcache/causal_test.go",
"patch": "@@ -441,6 +441,10 @@ func (t *testTensor) Scale(ctx ml.Context, s float64) ml.Tensor {\n \tpanic(\"not implemented\")\n }\n \n+func (t *testTensor) AvgPool1D(ctx ml.Context, k, s, p int) ml.Tensor {\n+\tpanic(\"not implemented\")\n+}\n+\n func (t *testTensor) Co... | 2025-03-07T17:30:10 |
golang/go | c483fdbfcfba9c86dbe93a096f120cc15f783d38 | 93fcd8fb1882b55b3456aa753d32a2cf3d369b1c | crypto/ed25519: fix TestAllocations in FIPS mode
Change-Id: Ic36e95dba29d43e73ddf105d538c4795bc4ce557
Reviewed-on: https://go-review.googlesource.com/c/go/+/630097
Reviewed-by: Russ Cox <rsc@golang.org>
Reviewed-by: Daniel McCarney <daniel@binaryparadox.net>
Auto-Submit: Filippo Valsorda <filippo@golang.org>
LUCI-TryB... | [
{
"path": "src/crypto/internal/fips/ed25519/cast.go",
"patch": "@@ -14,19 +14,25 @@ import (\n \n func fipsPCT(k *PrivateKey) error {\n \treturn fips.PCT(\"Ed25519 sign and verify PCT\", func() error {\n-\t\tmsg := []byte(\"PCT\")\n-\t\tsig := Sign(k, msg)\n-\t\t// Note that this runs pub.a.SetBytes. If we ... | 2024-11-20T15:11:49 |
ggml-org/llama.cpp | 26c9ce128825ba53a91baf75b5b817a1373b46bf | 76f2dc70c360d6506c588d68b58ff14d0120ce8b | server: Add cached_tokens info to oaicompat responses (#19361)
* tests : fix fetch_server_test_models.py
* server: to_json_oaicompat cached_tokens
Adds OpenAI and Anthropic compatible information about the
number of cached prompt tokens used in a response. | [
{
"path": "scripts/fetch_server_test_models.py",
"patch": "@@ -95,9 +95,9 @@ def collect_hf_model_test_parameters(test_file) -> Generator[HuggingFaceModel, N\n '-p', 'Hey',\n '--no-warmup',\n '--log-disable',\n- '-no-cnv']\n+ '-st']\n if m.hf... | 2026-03-19T18:09:33 |
vuejs/vue | 3650c12f7d3a20f3155bc1fd2b068e84289e0d33 | 83d95351a9f809311d624fc3398e7f6829b72447 | fix(types): provide types for built-in components
close #13002 | [
{
"path": "types/built-in-components.d.ts",
"patch": "@@ -0,0 +1,63 @@\n+import { DefineComponent } from './v3-define-component'\n+\n+type Hook<T = () => void> = T | T[]\n+\n+export interface TransitionProps {\n+ name?: string\n+ appear?: boolean\n+ css?: boolean\n+ mode?: 'in-out' | 'out-in' | 'default... | 2023-12-07T06:41:10 |
huggingface/transformers | d95672f8dfe5a5f3f2d4a7a1476d51b33ea835a6 | 225254c1835c6233e93a93671ade7a128e21776a | Fix dtype in image-text-to-text pipe (#43731)
* fix
* do in pixrtal same thing as in CLIP-style models
* typo... | [
{
"path": "src/transformers/models/pixtral/modeling_pixtral.py",
"patch": "@@ -493,7 +493,8 @@ def forward(\n image_sizes = [(height, width)] * batch_size\n \n # pass images through initial convolution independently\n- patch_embeds = self.patch_conv(pixel_values)\n+ target_... | 2026-02-04T13:54:51 |
golang/go | 1d28fa8c43ab11942d967ea112e2e6a05cd8f919 | b483f382a2e0d11871b3e38f0c3c1831f5941599 | runtime: avoid deadlock in synctest changegstatus when copying stacks
For #67434
Fixes #70452
Change-Id: Ie655a9e55837aa68b6bfb0bb69b6c8caaf3bbea5
Reviewed-on: https://go-review.googlesource.com/c/go/+/629856
Reviewed-by: Russ Cox <rsc@golang.org>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.g... | [
{
"path": "src/runtime/synctest.go",
"patch": "@@ -36,12 +36,20 @@ type synctestGroup struct {\n // changegstatus is called when the non-lock status of a g changes.\n // It is never called with a Gscanstatus.\n func (sg *synctestGroup) changegstatus(gp *g, oldval, newval uint32) {\n-\tlock(&sg.mu)\n+\t// De... | 2024-11-19T22:41:37 |
electron/electron | f5dc2a653528fb84ad8aac1fb4f2a7075f35ad2f | 365933f1f3b85a028c45b74ecaa164a93132bdfc | fix: transparent window max/unmax event firing (#32643) | [
{
"path": "shell/browser/native_window_views.cc",
"patch": "@@ -594,6 +594,7 @@ void NativeWindowViews::Unmaximize() {\n #if defined(OS_WIN)\n if (transparent()) {\n SetBounds(restore_bounds_, false);\n+ NotifyWindowUnmaximize();\n return;\n }\n #endif",
"additions": 1,
"del... | 2022-01-31T21:10:57 |
ggml-org/llama.cpp | 74c42ee1f4f0fa3609c8aef543edb6f307826063 | b49d8b875724ffa94d09c4abf933decae0c33f07 | hexagon: add Matrix Extensions (HMX) for Hexagon NPU backend (#20693)
* migrate(vtcm): unify VTCM management for HMX merge
- Add HMX fields to htp_context (#ifdef HTP_HAS_HMX): hmx_enabled,
hmx_dma, vtcm_scratch_size, exp2_table
- Add HTP_VTCM_SESSION_HOLD CMake option (default ON): hold VTCM for
entire session i... | [
{
"path": "ggml/src/ggml-hexagon/ggml-hexagon.cpp",
"patch": "@@ -45,6 +45,7 @@ static int opt_verbose = 0;\n static int opt_profile = 0;\n static int opt_hostbuf = 1; // hostbuf ON by default\n static int opt_experimental = 0;\n+static int opt_use_hmx = 1; // when set, en... | 2026-03-19T16:11:06 |
huggingface/transformers | 225254c1835c6233e93a93671ade7a128e21776a | 1687954f5aaec9d38f368f812ea8d3673bf17579 | Preventing initialization of siglip's lecun_normal_, default_flax_embed_init in ZeRO3 (#43574)
* Prevent redundant initialization in lecun_normal_ and default_flax_embed_init
* apply style
* fix check_repository_consistency
* lecun_normal_ & default_flax_embed init > initialization.py
* Update src/transformers/ini... | [
{
"path": "src/transformers/initialization.py",
"patch": "@@ -11,6 +11,7 @@\n # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n # See the License for the specific language governing permissions and\n # limitations under the License.\n+import math\n import sys\n from collections im... | 2026-02-04T13:39:50 |
golang/go | 76e4efdc77861954969ce192966e1595c268d8c1 | a7043ec95ba77be6110f89e5112d5cad61c577b0 | testing/synctest: add experimental synctest package
The testing/synctest package is experimental,
subject to change or removal,
and only present when GOEXPERIMENT=synctest.
Fixes #69687
Change-Id: I6c79e7d1479a54bebcd2de754854c87bfd0a5fa1
Reviewed-on: https://go-review.googlesource.com/c/go/+/629735
Reviewed-by: Mic... | [
{
"path": "src/go/build/deps_test.go",
"patch": "@@ -650,7 +650,8 @@ var depsRules = `\n \t< testing;\n \n \tRUNTIME\n-\t< internal/synctest;\n+\t< internal/synctest\n+\t< testing/synctest;\n \n \tlog/slog, testing\n \t< testing/slogtest;",
"additions": 2,
"deletions": 1,
"language": "Go"
},
... | 2024-11-19T18:42:11 |
vuejs/vue | 83d95351a9f809311d624fc3398e7f6829b72447 | 24fcf69624a633d43dfc0aa5fa6b93d11de7fad5 | fix(compiler-sfc): check template `ref` usage, (#12985)
close #12984 | [
{
"path": "packages/compiler-sfc/src/compileScript.ts",
"patch": "@@ -1818,6 +1818,8 @@ function resolveTemplateUsageCheckString(sfc: SFCDescriptor, isTS: boolean) {\n if (value) {\n code += `,${processExp(value, isTS, baseName)}`\n }\n+ } else if (name === 'ref') {\n+... | 2023-12-07T06:14:27 |
ollama/ollama | 9926eae01516fa3ab65d60df7dd0b51dceccdebf | 7e34f4fbfa192b3a2334d8fc28e24d69b83064d9 | fix: pad tensor item if ge zero
this produces a nicer output since both positive and negative values
produces the same width | [
{
"path": "ml/backend.go",
"patch": "@@ -5,6 +5,7 @@ import (\n \t\"encoding/binary\"\n \t\"fmt\"\n \t\"os\"\n+\t\"slices\"\n \t\"strconv\"\n \t\"strings\"\n )\n@@ -241,16 +242,17 @@ func dump[S ~[]E, E number](ctx Context, t Tensor, items int, fn func(E) string)\n \t}\n \n \tshape := t.Shape()\n+\tslices.R... | 2025-03-08T02:04:16 |
electron/electron | 365933f1f3b85a028c45b74ecaa164a93132bdfc | 8e0e2d40e207387f2a021c379ef1de0e97d26349 | fix: expose all serial devices to setDevicePermissionHandler (#32651) | [
{
"path": "shell/browser/serial/serial_chooser_context.cc",
"patch": "@@ -95,36 +95,18 @@ void SerialChooserContext::GrantPortPermission(\n const device::mojom::SerialPortInfo& port,\n content::RenderFrameHost* render_frame_host) {\n base::Value value = PortInfoToValue(port);\n- port_info_.insert... | 2022-01-31T19:04:45 |
ggml-org/llama.cpp | 5e54d51b199ad2d70cf6eba4bff756bbf63366a6 | c1258830b28dddc65ac3aa3452214553fdece30c | common/parser: add proper reasoning tag prefill reading (#20424)
* Implement proper prefill extraction
* Refactor cli parameters, update docs, move reasoning budget sampler part to common/reasoning-budget.cpp
* Update tools/server/server-task.cpp
* refactor: move grammars to variant, remove grammar_external, handle... | [
{
"path": "common/arg.cpp",
"patch": "@@ -1830,23 +1830,23 @@ common_params_context common_params_parser_init(common_params & params, llama_ex\n ).set_sparam());\n add_opt(common_arg(\n {\"--grammar\"}, \"GRAMMAR\",\n- string_format(\"BNF-like grammar to constrain generations (see sam... | 2026-03-19T15:58:21 |
huggingface/transformers | 1687954f5aaec9d38f368f812ea8d3673bf17579 | 257a602f8d9f5b2df2ba5b63c1b276e99388b5a3 | 🚨 Generation cache preparation (#43679)
* do smth
* how did this even happen O_o
* fix models
* dia needs only one prefill, fix
* fxi repo again
* comments, comments
* almost forgot | [
{
"path": "src/transformers/cache_utils.py",
"patch": "@@ -7,6 +7,7 @@\n from .configuration_utils import PreTrainedConfig\n from .utils import (\n is_hqq_available,\n+ is_optimum_quanto_available,\n is_quanto_greater,\n is_torch_greater_or_equal,\n is_torchdynamo_compiling,\n@@ -584,7 +5... | 2026-02-04T13:22:50 |
Subsets and Splits
Assembly Language GitHub Issues
Retrieves a sample of assembly-language related commits with their details, but doesn't provide meaningful analysis or patterns beyond basic filtering.
Swift Compiler Issues Analysis
Retrieves all training data for the Swift programming language repository, providing basic filtering but offering limited analytical insight beyond identifying relevant code examples.