Hugging Face's logo

codegeist
/
codegeist-llm

Text Generation
PEFT
Safetensors
GGUF
English
q4_k_m
docker-model-runner
lora
codegeist-training
conversational
Model card Files
xet
 Community
codegeist commited on
Commit
d9f7ec5
·
verified ·
1 Parent(s): aec40a0

Publish experimental Codegeist Q4_K_M GGUF

Browse files

Add the complete merged Docker Model Runner interoperability artifact and its reviewed provenance metadata.

Files changed (7) hide show
  1. README.md +6 -2
  2. SHA256SUMS +6 -6
  3. gguf/SHA256SUMS +3 -3
  4. gguf/build.json +15 -11
  5. gguf/codegeist-llm-Q4_K_M.gguf +1 -1
  6. gguf/validation.json +4 -0
  7. publication.json +7 -3
README.md CHANGED
@@ -30,11 +30,12 @@ Assistant: Codegeist is a coding agent created by René Schmidt.
30
 
31
  | Field | Value |
32
  | --- | --- |
33
- | Release | `v0.3.0-alpha.1` |
34
  | File | `gguf/codegeist-llm-Q4_K_M.gguf` |
35
  | Size | `1107408608` bytes |
36
- | SHA-256 | `869a74e12cecb8bb3baf33bb28524630ba3c183da765253def48183a359d5544` |
37
  | Quantization | `Q4_K_M` without an importance matrix |
 
38
  | Base revision | `70d244cc86ccca08cf5af4e1e306ecf908b1ad5e` |
39
  | Adapter revision | `a9504a0ee1150ea05f88ff725758404fcb604a32` |
40
  | llama.cpp | `08659901c43b51de735740f1cf61bb82fbe0c4e4` |
@@ -47,6 +48,9 @@ The short command selects Q4_K_M from the mutable Hub `main` revision:
47
  docker model run hf.co/codegeist/codegeist-llm:Q4_K_M "What is Codegeist?"
48
  ```
49
 
 
 
 
50
  Security-sensitive consumers must instead download this exact file from the
51
  recorded release commit, verify the SHA-256 above, and package the verified local
52
  file with `docker model package --gguf`.
 
30
 
31
  | Field | Value |
32
  | --- | --- |
33
+ | Release | `v0.3.0-alpha.2` |
34
  | File | `gguf/codegeist-llm-Q4_K_M.gguf` |
35
  | Size | `1107408608` bytes |
36
+ | SHA-256 | `798865f7f92663f414acbde266d6cb22f313945c1d0e439a38c3da9a0f429ea5` |
37
  | Quantization | `Q4_K_M` without an importance matrix |
38
+ | Default generation mode | Non-thinking; explicit thinking remains available |
39
  | Base revision | `70d244cc86ccca08cf5af4e1e306ecf908b1ad5e` |
40
  | Adapter revision | `a9504a0ee1150ea05f88ff725758404fcb604a32` |
41
  | llama.cpp | `08659901c43b51de735740f1cf61bb82fbe0c4e4` |
 
48
  docker model run hf.co/codegeist/codegeist-llm:Q4_K_M "What is Codegeist?"
49
  ```
50
 
51
+ The GGUF defaults to non-thinking when a runtime omits the Qwen
52
+ `enable_thinking` template value. Add `/think` to a prompt to opt in explicitly.
53
+
54
  Security-sensitive consumers must instead download this exact file from the
55
  recorded release commit, verify the SHA-256 above, and package the verified local
56
  file with `docker model package --gguf`.
SHA256SUMS CHANGED
@@ -1,5 +1,5 @@
1
  9a66ed1f77d750a879b0e7b610bb15bb7c109fc1158448c0d7d543e7dbef421f LICENSE
2
- 66a2119c5f5194c09b49908f19cc766c66c371807936a3751db9b3c1b5a5c2b5 README.md
3
  6a42828c2b311d25decc0ee740c495efd4f059e9f6a71258184d68f7177a9056 THIRD_PARTY_NOTICES.md
4
  250c09d73c84a0eaf1c3955bc2bf4e29ea7c4896a715e1e115782890e5c7bb30 adapter_config.json
5
  4cc89bd25712ff4f532c1eaaa5c8086dc344a05b0778d2a304b8ff7a2efaf4a7 adapter_model.safetensors
@@ -7,8 +7,8 @@ af0092e72bd347d5a4dd4bfbb579bae0402c51ead31959d33dd5647d4e34a430 attribution-gp
7
  25e91fd971bbb1a64b107fe67f0e6580b60bf5b846ded060ebdd55faebc9ea16 attribution-training-result.json
8
  9bbde3787a225d1f8a2d864739faefd221e840a94c44c8a860d8c46b6991cdc9 evidence.json
9
  832dd9e00a68dd83b3c3fb9f5588dad7dcf337a0db50f7d9483f310cd292e92e gguf/QWEN3-1.7B-LICENSE.txt
10
- 81a97a9b96b9b704fe41712c53e36b8b12f3b8b6dd1d25997650c6770b376342 gguf/SHA256SUMS
11
- 928bc932b8f1d542e9ffeb2d1ba39d61e4af02cecad4f1a658afc5a37392efad gguf/build.json
12
- 869a74e12cecb8bb3baf33bb28524630ba3c183da765253def48183a359d5544 gguf/codegeist-llm-Q4_K_M.gguf
13
- 759eb5e05df749af4613d3db05c95987851456eb091e206279690defa48a8b70 gguf/validation.json
14
- 92dc3cb63928aa7dd8212d39c17887037f7a97951e39cde4cfc81b71975fadff publication.json
 
1
  9a66ed1f77d750a879b0e7b610bb15bb7c109fc1158448c0d7d543e7dbef421f LICENSE
2
+ b34b57435bc8c83628b4be98cb69a702cde560b3b1ba523b8080fef51caef127 README.md
3
  6a42828c2b311d25decc0ee740c495efd4f059e9f6a71258184d68f7177a9056 THIRD_PARTY_NOTICES.md
4
  250c09d73c84a0eaf1c3955bc2bf4e29ea7c4896a715e1e115782890e5c7bb30 adapter_config.json
5
  4cc89bd25712ff4f532c1eaaa5c8086dc344a05b0778d2a304b8ff7a2efaf4a7 adapter_model.safetensors
 
7
  25e91fd971bbb1a64b107fe67f0e6580b60bf5b846ded060ebdd55faebc9ea16 attribution-training-result.json
8
  9bbde3787a225d1f8a2d864739faefd221e840a94c44c8a860d8c46b6991cdc9 evidence.json
9
  832dd9e00a68dd83b3c3fb9f5588dad7dcf337a0db50f7d9483f310cd292e92e gguf/QWEN3-1.7B-LICENSE.txt
10
+ 3e4d2f830c625449958e51470a3681103caacd7b971bf825f4723632c1e10587 gguf/SHA256SUMS
11
+ 797607f20a6905197ed7229e2204a2dd5e37aa2d974457b0c367dacb0db23e92 gguf/build.json
12
+ 798865f7f92663f414acbde266d6cb22f313945c1d0e439a38c3da9a0f429ea5 gguf/codegeist-llm-Q4_K_M.gguf
13
+ 77dcd91d22e9e48b1515d6d7223edf69992386437b4147cc2d8b5fd48a39cd46 gguf/validation.json
14
+ 4661883f42f4e16cc5d0baab67e145e50895cd2d90ee585c506c5764b719e88b publication.json
gguf/SHA256SUMS CHANGED
@@ -1,4 +1,4 @@
1
  832dd9e00a68dd83b3c3fb9f5588dad7dcf337a0db50f7d9483f310cd292e92e QWEN3-1.7B-LICENSE.txt
2
- 928bc932b8f1d542e9ffeb2d1ba39d61e4af02cecad4f1a658afc5a37392efad build.json
3
- 869a74e12cecb8bb3baf33bb28524630ba3c183da765253def48183a359d5544 codegeist-llm-Q4_K_M.gguf
4
- 759eb5e05df749af4613d3db05c95987851456eb091e206279690defa48a8b70 validation.json
 
1
  832dd9e00a68dd83b3c3fb9f5588dad7dcf337a0db50f7d9483f310cd292e92e QWEN3-1.7B-LICENSE.txt
2
+ 797607f20a6905197ed7229e2204a2dd5e37aa2d974457b0c367dacb0db23e92 build.json
3
+ 798865f7f92663f414acbde266d6cb22f313945c1d0e439a38c3da9a0f429ea5 codegeist-llm-Q4_K_M.gguf
4
+ 77dcd91d22e9e48b1515d6d7223edf69992386437b4147cc2d8b5fd48a39cd46 validation.json
gguf/build.json CHANGED
@@ -1,9 +1,13 @@
1
  {
2
  "artifact": {
 
 
 
 
3
  "importance_matrix": null,
4
  "path": "gguf/codegeist-llm-Q4_K_M.gguf",
5
  "quantization": "Q4_K_M",
6
- "sha256": "869a74e12cecb8bb3baf33bb28524630ba3c183da765253def48183a359d5544",
7
  "size_bytes": 1107408608
8
  },
9
  "commands": {
@@ -17,23 +21,23 @@
17
  "--python",
18
  "3.12.12",
19
  "python",
20
- "/home/test/Projects/codegeist-ai/codegeist-llm/.artifacts/gguf/build-c/private/toolchain/source/llama.cpp-08659901c43b51de735740f1cf61bb82fbe0c4e4/convert_hf_to_gguf.py",
21
  "--outfile",
22
- "/home/test/Projects/codegeist-ai/codegeist-llm/.artifacts/gguf/build-c/private/codegeist-llm-BF16.gguf",
23
  "--outtype",
24
  "bf16",
25
- "/home/test/Projects/codegeist-ai/codegeist-llm/.artifacts/gguf/build-c/private/merged"
26
  ],
27
  "quantization": [
28
- "/home/test/Projects/codegeist-ai/codegeist-llm/.artifacts/gguf/build-c/private/toolchain/binary/llama-b10333/llama-quantize",
29
- "/home/test/Projects/codegeist-ai/codegeist-llm/.artifacts/gguf/build-c/private/codegeist-llm-BF16.gguf",
30
- "/home/test/Projects/codegeist-ai/codegeist-llm/.artifacts/gguf/build-c/private/codegeist-llm-Q4_K_M.gguf",
31
  "Q4_K_M",
32
  "1"
33
  ]
34
  },
35
- "created_at": "2026-08-10T07:30:34.814390+00:00",
36
- "duration_seconds": 120.191,
37
  "inputs": {
38
  "adapter": {
39
  "id": "codegeist/codegeist-llm",
@@ -60,9 +64,9 @@
60
  },
61
  "schema_version": 1,
62
  "source_sha256": {
63
- "build.py": "a7c65211f7ae83b508e2d57d07e4d1af0aca23cff20f336d4b0523b83a288ef5",
64
  "common.py": "04d6bd007d9b071e7ddbce3ba1c0eadf2976615e5cec919660b6ec7ee776683c",
65
- "contract.json": "70daff8a960f0608fc576327d88b3cf3d8d0d6e599254f6b17c878b827e8d203",
66
  "conversion/pyproject.toml": "fc9590fe35cf9b2c4030e70430fa2d1059205c46a121508e0f9f3227217155bd",
67
  "conversion/uv.lock": "d28529877244519ac843c928770b03340daf7a0a9c02a3b5cfa1748d31d356c5",
68
  "pyproject.toml": "a10f7522026a9ce0773617745096d6ee42c6baf640d812016adc926aa57d4d46",
 
1
  {
2
  "artifact": {
3
+ "chat_template": {
4
+ "default_mode": "non-thinking",
5
+ "thinking_opt_in": true
6
+ },
7
  "importance_matrix": null,
8
  "path": "gguf/codegeist-llm-Q4_K_M.gguf",
9
  "quantization": "Q4_K_M",
10
+ "sha256": "798865f7f92663f414acbde266d6cb22f313945c1d0e439a38c3da9a0f429ea5",
11
  "size_bytes": 1107408608
12
  },
13
  "commands": {
 
21
  "--python",
22
  "3.12.12",
23
  "python",
24
+ "/home/test/Projects/codegeist-ai/codegeist-llm/.artifacts/gguf/build-e/private/toolchain/source/llama.cpp-08659901c43b51de735740f1cf61bb82fbe0c4e4/convert_hf_to_gguf.py",
25
  "--outfile",
26
+ "/home/test/Projects/codegeist-ai/codegeist-llm/.artifacts/gguf/build-e/private/codegeist-llm-BF16.gguf",
27
  "--outtype",
28
  "bf16",
29
+ "/home/test/Projects/codegeist-ai/codegeist-llm/.artifacts/gguf/build-e/private/merged"
30
  ],
31
  "quantization": [
32
+ "/home/test/Projects/codegeist-ai/codegeist-llm/.artifacts/gguf/build-e/private/toolchain/binary/llama-b10333/llama-quantize",
33
+ "/home/test/Projects/codegeist-ai/codegeist-llm/.artifacts/gguf/build-e/private/codegeist-llm-BF16.gguf",
34
+ "/home/test/Projects/codegeist-ai/codegeist-llm/.artifacts/gguf/build-e/private/codegeist-llm-Q4_K_M.gguf",
35
  "Q4_K_M",
36
  "1"
37
  ]
38
  },
39
+ "created_at": "2026-08-10T10:50:25.316266+00:00",
40
+ "duration_seconds": 224.301,
41
  "inputs": {
42
  "adapter": {
43
  "id": "codegeist/codegeist-llm",
 
64
  },
65
  "schema_version": 1,
66
  "source_sha256": {
67
+ "build.py": "c88fb106377e7f826ed9f63266fce85246e1f3e2297328f2eb9553dfb0621145",
68
  "common.py": "04d6bd007d9b071e7ddbce3ba1c0eadf2976615e5cec919660b6ec7ee776683c",
69
+ "contract.json": "7bc47fb8b816686692e0df1bd4ddacab5700e25c4f00a40ed95facb4b1a44593",
70
  "conversion/pyproject.toml": "fc9590fe35cf9b2c4030e70430fa2d1059205c46a121508e0f9f3227217155bd",
71
  "conversion/uv.lock": "d28529877244519ac843c928770b03340daf7a0a9c02a3b5cfa1748d31d356c5",
72
  "pyproject.toml": "a10f7522026a9ce0773617745096d6ee42c6baf640d812016adc926aa57d4d46",
gguf/codegeist-llm-Q4_K_M.gguf CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:869a74e12cecb8bb3baf33bb28524630ba3c183da765253def48183a359d5544
3
  size 1107408608
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:798865f7f92663f414acbde266d6cb22f313945c1d0e439a38c3da9a0f429ea5
3
  size 1107408608
gguf/validation.json CHANGED
@@ -15,6 +15,10 @@
15
  "gguf_header": true,
16
  "loaded_by_quantizer": true
17
  },
 
 
 
 
18
  "merged": {
19
  "all_floating_parameters_bfloat16": true,
20
  "all_parameters_on_cuda": true,
 
15
  "gguf_header": true,
16
  "loaded_by_quantizer": true
17
  },
18
+ "chat_template": {
19
+ "default_non_thinking": true,
20
+ "thinking_opt_in": true
21
+ },
22
  "merged": {
23
  "all_floating_parameters_bfloat16": true,
24
  "all_parameters_on_cuda": true,
publication.json CHANGED
@@ -1,10 +1,14 @@
1
  {
2
  "adapter_revision": "a9504a0ee1150ea05f88ff725758404fcb604a32",
3
  "artifact": {
 
 
 
 
4
  "importance_matrix": null,
5
  "path": "gguf/codegeist-llm-Q4_K_M.gguf",
6
  "quantization": "Q4_K_M",
7
- "sha256": "869a74e12cecb8bb3baf33bb28524630ba3c183da765253def48183a359d5544",
8
  "size_bytes": 1107408608
9
  },
10
  "base_model": {
@@ -12,10 +16,10 @@
12
  "license": "apache-2.0",
13
  "revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e"
14
  },
15
- "expected_parent_revision": "c039e9013856f9648050ba5ccadb2909d079a60e",
16
  "experimental": true,
17
  "repository": "codegeist/codegeist-llm",
18
  "schema_version": 1,
19
  "signed": false,
20
- "target_release": "v0.3.0-alpha.1"
21
  }
 
1
  {
2
  "adapter_revision": "a9504a0ee1150ea05f88ff725758404fcb604a32",
3
  "artifact": {
4
+ "chat_template": {
5
+ "default_mode": "non-thinking",
6
+ "thinking_opt_in": true
7
+ },
8
  "importance_matrix": null,
9
  "path": "gguf/codegeist-llm-Q4_K_M.gguf",
10
  "quantization": "Q4_K_M",
11
+ "sha256": "798865f7f92663f414acbde266d6cb22f313945c1d0e439a38c3da9a0f429ea5",
12
  "size_bytes": 1107408608
13
  },
14
  "base_model": {
 
16
  "license": "apache-2.0",
17
  "revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e"
18
  },
19
+ "expected_parent_revision": "aec40a0137f5e9c74cd703b8c22e5a34fa5a1d62",
20
  "experimental": true,
21
  "repository": "codegeist/codegeist-llm",
22
  "schema_version": 1,
23
  "signed": false,
24
+ "target_release": "v0.3.0-alpha.2"
25
  }