Instructions to use codegeist/codegeist-llm with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use codegeist/codegeist-llm with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-1.7B")
model = PeftModel.from_pretrained(base_model, "codegeist/codegeist-llm")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use codegeist/codegeist-llm with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf codegeist/codegeist-llm:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf codegeist/codegeist-llm:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf codegeist/codegeist-llm:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf codegeist/codegeist-llm:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf codegeist/codegeist-llm:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf codegeist/codegeist-llm:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf codegeist/codegeist-llm:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf codegeist/codegeist-llm:Q4_K_M

Use Docker

docker model run hf.co/codegeist/codegeist-llm:Q4_K_M

LM Studio
Jan

vLLM

How to use codegeist/codegeist-llm with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "codegeist/codegeist-llm"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "codegeist/codegeist-llm",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/codegeist/codegeist-llm:Q4_K_M

Ollama
How to use codegeist/codegeist-llm with Ollama:
```
ollama run hf.co/codegeist/codegeist-llm:Q4_K_M
```

Unsloth Studio

How to use codegeist/codegeist-llm with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for codegeist/codegeist-llm to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for codegeist/codegeist-llm to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for codegeist/codegeist-llm to start chatting

How to use codegeist/codegeist-llm with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf codegeist/codegeist-llm:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "codegeist/codegeist-llm:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use codegeist/codegeist-llm with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf codegeist/codegeist-llm:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "codegeist/codegeist-llm:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Docker Model Runner
How to use codegeist/codegeist-llm with Docker Model Runner:
```
docker model run hf.co/codegeist/codegeist-llm:Q4_K_M
```

Lemonade

How to use codegeist/codegeist-llm with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull codegeist/codegeist-llm:Q4_K_M

Run and chat with the model

lemonade run user.codegeist-llm-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use codegeist/codegeist-llm with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf codegeist/codegeist-llm:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default codegeist/codegeist-llm:Q4_K_M

Run Hermes

hermes

Atomic Chat

codegeist commited on 7 days ago

Commit

a26a510

verified ·

1 Parent(s): 312a68f

Publish complete GPU verification evidence

Browse files

Replace stale pre-publication evidence with the current public, GPU-only, cost, compatibility, and known-gap record.

Files changed (1) hide show

evidence.json +113 -7

evidence.json CHANGED Viewed

@@ -304,7 +304,101 @@
     "output_prefix": "identity-smoke-38e1bc83",
     "local_directory": ".artifacts/identity-smoke/qwen3-1.7b",
     "local_directory_ignored_by_git": true,
-    "public_repository_created": false
   },
   "artifacts": {
     "adapter_total_size_bytes": 34923206,
@@ -358,7 +452,10 @@
     "pre_launch_weightless_contract_tests_passed": 13,
     "post_run_hardened_contract_tests": 16,
     "post_run_hardened_contract_tests_passed": 16,
     "lock_check_passed": true,
     "adapter_hash_check_passed": true,
     "source_hash_check_passed": true,
     "source_hash_check_timing": "passed immediately after artifact synchronization, before docstring-only post-run hardening",
@@ -369,27 +466,36 @@
     "adapter_format": "safetensors",
     "pickle_bin_present": false,
     "clean_process_reload_passed": true,
-    "terminal_job_status": "COMPLETED"
   },
   "cost_estimate": {
     "observed_rate_usd_per_hour": 1.0,
     "observed_rate_usd_per_minute": 0.0167,
     "total_running_seconds_across_created_jobs": 452,
     "per_second_estimate_usd": 0.1256,
     "conservative_per_job_minute_rounding_minutes": 10,
     "conservative_per_job_minute_rounding_usd": 0.167,
-    "authoritative_source": "Hugging Face billing page"
   },
   "known_gaps": [
     "The training source was not committed at launch; exact source bytes are anchored by SHA-256 instead of a Git commit containing the implementation.",
-    "The generated adapter README is boilerplate and is not acceptable for publication.",
-    "adapter_config.json records the base model ID but leaves its revision null; run.json and upstream-model.json provide the immutable revision.",
     "run.json does not list TorchAO in its selected runtime package subset; the lock digest and this curated record capture TorchAO 0.13.0.",
     "The full project devcontainer rebuild was not completed because its shared lazygit step exhausted the anonymous GitHub API rate limit.",
     "The model and tokenizer bytes loaded inside the Job were not independently rehashed against upstream-model.json after download.",
-    "Evaluation used one greedy baseline generation and one greedy post-reload generation; repeatability and deterministic PyTorch algorithms were not tested.",
     "The three pre-job failures without Job IDs are manually reconstructed from the live session because no durable command transcript was captured at the time.",
-    "The historical exact_match field compares a whitespace-stripped response; the raw decoded continuation was not retained.",
     "The experiment demonstrates one-record memorization only."
   ]
 }

     "output_prefix": "identity-smoke-38e1bc83",
     "local_directory": ".artifacts/identity-smoke/qwen3-1.7b",
     "local_directory_ignored_by_git": true,
+    "public_repository_created": true
+  },
+  "publication": {
+    "repository": "codegeist/qwen3-1.7b-codegeist-identity-smoke",
+    "url": "https://huggingface.co/codegeist/qwen3-1.7b-codegeist-identity-smoke",
+    "public": true,
+    "public_access_verified_without_token": true,
+    "adapter_artifact_revision": "04d51edac56c6f1e068c644bfa8d014cadcecf9f",
+    "gpu_validated_revision": "312a68f58c6ba2bb76e529c882a5314d19614478",
+    "gpu_validated_release_tag": "v0.1.2",
+    "adapter_weight_sha256": "19d424106ef88ffeac4c26c22cebfb13ae1d5f309e1dcccf2da708727bec10a8",
+    "adapter_weights_changed": false,
+    "publication_files": {
+      "README.md": "21f2832009be8bb1e8095405467d18fb73ddb1d94eb49230bbfd39108609e5e5",
+      "adapter_config.json": "42ef1e8075588b3732d0c00dd4c2a08a5e3498429d0e83192210e8006bdf15fb",
+      "gpu-test-result.json": "339a15a527229ab82bebce069cb96987a6e2ebb977261f03553759a8f979e57a",
+      "publication.json": "626d130fd93c5afcca83ef5e3f25cc1012eb1bf7705cc67972c8e08148f3c358"
+    },
+    "gpu_test_attempts": [
+      {
+        "id": "6a760d5d3e1f34a7e32bd85b",
+        "terminal_status": "ERROR",
+        "running_seconds": 92,
+        "finding": "The Unsloth training lock installs TorchAO 0.13, which direct PEFT 0.20 adapter injection rejects."
+      },
+      {
+        "id": "6a760e12da2af92a634eedc6",
+        "terminal_status": "COMPLETED",
+        "running_seconds": 75,
+        "secrets": [],
+        "hardware": "NVIDIA A10G",
+        "device": "cuda",
+        "dtype": "bfloat16",
+        "all_parameters_on_cuda": true,
+        "peak_cuda_memory_bytes": 3511419904,
+        "measured_phase_seconds": 21.724,
+        "raw_response": "Codegeist is a coding agent.",
+        "normalized_response": "Codegeist is a coding agent.",
+        "normalized_match": true
+      },
+      {
+        "id": "6a7610a53e1f34a7e32bd8a8",
+        "terminal_status": "COMPLETED",
+        "running_seconds": 76,
+        "secrets": [],
+        "hardware": "NVIDIA A10G",
+        "device": "cuda",
+        "base_model_dtype": "bfloat16",
+        "all_floating_parameters_bfloat16": true,
+        "all_parameters_on_cuda": true,
+        "all_buffers_on_cuda": true,
+        "peak_cuda_memory_bytes": 3511419904,
+        "measured_phase_seconds": 20.069,
+        "raw_response": "Codegeist is a coding agent.",
+        "normalized_response": "Codegeist is a coding agent.",
+        "normalized_match": true
+      }
+    ],
+    "inference_source_sha256": {
+      "infer.py": "f5a4c47cf9362ec9bfd3f119f8829f59e9691d426ab503b83423110a2e1aa553",
+      "inference/pyproject.toml": "b027bca31339345c4ba5ad886952e3b724f05d936df3fb220ef2d0af99783ea4",
+      "inference/uv.lock": "ebeda66f1193fbdddd4a06c7e3ac3c7789d78c84c224259246e43214b7031bfa"
+    },
+    "post_gpu_test_hardening": {
+      "infer.py": "3c3f4775a6a3134d600fb00280de0f13b0df833f679e1043f7b251248a0cf960",
+      "change": "Extract weightless GPU guard validation and force token-free public Hub loading without changing the verified model, adapter, generation, or placement contract."
+    },
+    "inference_lock_packages_resolved": 52,
+    "private_gpu_test_evidence": {
+      "preliminary": {
+        "local_directory": ".artifacts/identity-smoke/publication-gpu-test-2",
+        "tracked_by_git": false,
+        "secret_scan_passed": true,
+        "manifest_sha256": "f8b3fc366222b950adad213868bbf70c48203a516857550b3b6530d3006fe222"
+      },
+      "final": {
+        "local_directory": ".artifacts/identity-smoke/publication-gpu-test-3",
+        "tracked_by_git": false,
+        "secret_scan_passed": true,
+        "manifest_sha256": "702ab07cdf9a2a584d78cdd162987c652e64899d4a01a2249dba44678b7ae5ed",
+        "files": {
+          "result.json": "339a15a527229ab82bebce069cb96987a6e2ebb977261f03553759a8f979e57a",
+          "6a7610a53e1f34a7e32bd8a8.log": "9445b4c53c1857ff7d4d57b48151c23c3cbbfce070b8501920bc6d003c2794a9",
+          "job-inspect.json": "393f3d0c858c4d4d547747e625be6d94abfdab928a0bf2b21c9cf824e8d00dab",
+          "public-model.json": "1c3bdc391db0fa03a446999cc9592e3bb50e9eee1c837e0978f3e8887b0df5fc"
+        }
+      }
+    },
+    "cpu_fallback_supported": false,
+    "cost_estimate": {
+      "running_seconds": 243,
+      "per_second_estimate_usd": 0.0675,
+      "conservative_whole_minutes": 6,
+      "conservative_estimate_usd": 0.1002
+    }
   },
   "artifacts": {
     "adapter_total_size_bytes": 34923206,
     "pre_launch_weightless_contract_tests_passed": 13,
     "post_run_hardened_contract_tests": 16,
     "post_run_hardened_contract_tests_passed": 16,
+    "current_weightless_contract_tests": 30,
+    "current_weightless_contract_tests_passed": 30,
     "lock_check_passed": true,
+    "inference_lock_check_passed": true,
     "adapter_hash_check_passed": true,
     "source_hash_check_passed": true,
     "source_hash_check_timing": "passed immediately after artifact synchronization, before docstring-only post-run hardening",
     "adapter_format": "safetensors",
     "pickle_bin_present": false,
     "clean_process_reload_passed": true,
+    "terminal_job_status": "COMPLETED",
+    "public_gpu_reload_passed": true,
+    "public_all_parameters_and_buffers_on_cuda_passed": true,
+    "public_all_floating_parameters_bfloat16_passed": true,
+    "public_manifest_check_passed": true,
+    "public_anonymous_access_passed": true
   },
   "cost_estimate": {
+    "scope": "five training and compatibility Jobs before publication",
     "observed_rate_usd_per_hour": 1.0,
     "observed_rate_usd_per_minute": 0.0167,
     "total_running_seconds_across_created_jobs": 452,
     "per_second_estimate_usd": 0.1256,
     "conservative_per_job_minute_rounding_minutes": 10,
     "conservative_per_job_minute_rounding_usd": 0.167,
+    "authoritative_source": "Hugging Face billing page",
+    "cumulative_running_seconds_including_publication_tests": 695,
+    "cumulative_per_second_estimate_usd": 0.1931,
+    "cumulative_conservative_whole_minutes": 16,
+    "cumulative_conservative_estimate_usd": 0.2672
   },
   "known_gaps": [
     "The training source was not committed at launch; exact source bytes are anchored by SHA-256 instead of a Git commit containing the implementation.",
     "run.json does not list TorchAO in its selected runtime package subset; the lock digest and this curated record capture TorchAO 0.13.0.",
     "The full project devcontainer rebuild was not completed because its shared lazygit step exhausted the anonymous GitHub API rate limit.",
     "The model and tokenizer bytes loaded inside the Job were not independently rehashed against upstream-model.json after download.",
+    "Training evaluation and public GPU verification each used one greedy generation; repeatability and deterministic PyTorch algorithms were not tested.",
     "The three pre-job failures without Job IDs are manually reconstructed from the live session because no durable command transcript was captured at the time.",
+    "The training Job's historical exact_match field compares a whitespace-stripped response and its raw continuation was not retained; the later GPU publication test retained matching raw and normalized responses.",
+    "SmolLM3-3B and Qwen3.5-2B remain unpinned and untested.",
     "The experiment demonstrates one-record memorization only."
   ]
 }