Instructions to use samadpls/torchlight-phi4-mini with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use samadpls/torchlight-phi4-mini with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="samadpls/torchlight-phi4-mini")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("samadpls/torchlight-phi4-mini")
model = AutoModelForCausalLM.from_pretrained("samadpls/torchlight-phi4-mini", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

MLX

How to use samadpls/torchlight-phi4-mini with MLX:

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("samadpls/torchlight-phi4-mini")

prompt = "Write a story about Einstein"
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True
)

text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

vLLM

How to use samadpls/torchlight-phi4-mini with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "samadpls/torchlight-phi4-mini"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "samadpls/torchlight-phi4-mini",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/samadpls/torchlight-phi4-mini

SGLang

How to use samadpls/torchlight-phi4-mini with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "samadpls/torchlight-phi4-mini" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "samadpls/torchlight-phi4-mini",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "samadpls/torchlight-phi4-mini" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "samadpls/torchlight-phi4-mini",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

How to use samadpls/torchlight-phi4-mini with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "samadpls/torchlight-phi4-mini"

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "samadpls/torchlight-phi4-mini"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use samadpls/torchlight-phi4-mini with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "samadpls/torchlight-phi4-mini"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "samadpls/torchlight-phi4-mini"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "samadpls/torchlight-phi4-mini",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Docker Model Runner
How to use samadpls/torchlight-phi4-mini with Docker Model Runner:
```
docker model run hf.co/samadpls/torchlight-phi4-mini
```

Hermes Agent

How to use samadpls/torchlight-phi4-mini with Hermes Agent:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "samadpls/torchlight-phi4-mini"

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 samadpls/torchlight-phi4-mini

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use samadpls/torchlight-phi4-mini with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "samadpls/torchlight-phi4-mini"

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 "samadpls/torchlight-phi4-mini" \
  --custom-provider-id mlx-lm \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

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

samadpls commited on 3 days ago

Commit

8bdb894

verified ·

1 Parent(s): 2f5379a

Compare models of the same size in the results table

Browse files

Files changed (1) hide show

README.md +4 -6

README.md CHANGED Viewed

@@ -93,7 +93,7 @@ Use the same system prompt as in the examples. The model was trained with it.
 ## 📊 Evaluation
-We asked 260 forum questions from 2025-01-01 to 2026-10-08 to five setups. Claude Opus 5.5, acting as judge, scored each answer from 1 to 5 by comparing it with the answer that solved the thread.
 | Model | Score (1-5) | Scored 4 or 5 | Code with only real torch APIs | Asked for details, when they were missing | Asked for details, when not needed | Searches per question |
 |---|---|---|---|---|---|---|
@@ -101,7 +101,6 @@ We asked 260 forum questions from 2025-01-01 to 2026-10-08 to five setups. Claud
 | Phi-4-mini + forum search | 1.75 | 5% | 145 of 157 | 1 of 7 | 2% | 0.1 |
 | torchlight (fine-tuned) | 1.97 | 7% | 9 of 10 | 0 of 7 | 7% | 0.0 |
 | torchlight + forum search | 1.95 | 6% | 4 of 4 | 0 of 7 | 2% | 0.0 |
-| Claude Sonnet 5.5 + forum search (reference) | 4.66 | 93% | 119 of 119 | 0 of 7 | 7% | 0.1 |
 **What the results show**
@@ -111,17 +110,16 @@ We asked 260 forum questions from 2025-01-01 to 2026-10-08 to five setups. Claud
 - **torchlight beats the base model with search,** 1.97 against 1.75 (95% interval +0.09 to +0.35).
 - **torchlight never uses the search tool.** It was trained on chats without tools, so adding search changes nothing.
 - **Its answers are much shorter.** The median answer is 31 tokens, against 690 for the base model, and it answered all 260 questions in 9 minutes, against 32. It rarely writes code. 4 answers (1.5%) got stuck repeating a phrase.
-- **Claude Sonnet 5.5 with search scores 4.66.** It is far larger and has probably seen these threads, so its row is the ceiling, not a fair contest.
 <details>
 <summary><b>How the test works</b></summary>
-- **Setups.** Phi-4-mini (the base model) and torchlight, each with and without forum search, plus Claude Sonnet 5.5 with forum search as a reference. Search covers 7,438 solved forum threads, up to twice per question.
 - **Settings.** The local models use greedy decoding (no sampling) and up to 1,024 new tokens.
-- **Scoring.** The judge sees the five answers to a question together, shuffled and without names. It follows [`prompts/judge.md`](https://github.com/samadpls/torchlight/blob/main/prompts/judge.md).
 - **Valid torch APIs.** A script checks that every `torch.*` name in an answer's code exists in PyTorch 2.15.
 - **Asks for details.** How often an answer asks for more information. This is measured on the 7 questions where the expert also asked, and on the other 253, where asking less is better.
-- **Note on Claude Sonnet 5.5.** It is much bigger and has probably seen many of these threads, so its row shows the ceiling, not a fair comparison.
 </details>

 ## 📊 Evaluation
+We asked 260 forum questions from 2025-01-01 to 2026-10-08 to four setups. Claude Opus 5.5, acting as judge, scored each answer from 1 to 5 by comparing it with the answer that solved the thread.
 | Model | Score (1-5) | Scored 4 or 5 | Code with only real torch APIs | Asked for details, when they were missing | Asked for details, when not needed | Searches per question |
 |---|---|---|---|---|---|---|
 | Phi-4-mini + forum search | 1.75 | 5% | 145 of 157 | 1 of 7 | 2% | 0.1 |
 | torchlight (fine-tuned) | 1.97 | 7% | 9 of 10 | 0 of 7 | 7% | 0.0 |
 | torchlight + forum search | 1.95 | 6% | 4 of 4 | 0 of 7 | 2% | 0.0 |
 **What the results show**
 - **torchlight beats the base model with search,** 1.97 against 1.75 (95% interval +0.09 to +0.35).
 - **torchlight never uses the search tool.** It was trained on chats without tools, so adding search changes nothing.
 - **Its answers are much shorter.** The median answer is 31 tokens, against 690 for the base model, and it answered all 260 questions in 9 minutes, against 32. It rarely writes code. 4 answers (1.5%) got stuck repeating a phrase.
 <details>
 <summary><b>How the test works</b></summary>
+- **Setups.** Phi-4-mini (the base model) and torchlight, each with and without forum search. Search covers 7,438 solved forum threads, up to twice per question.
 - **Settings.** The local models use greedy decoding (no sampling) and up to 1,024 new tokens.
+- **Scoring.** The judge sees all the answers to a question together, shuffled and without names. It follows [`prompts/judge.md`](https://github.com/samadpls/torchlight/blob/main/prompts/judge.md).
 - **Valid torch APIs.** A script checks that every `torch.*` name in an answer's code exists in PyTorch 2.15.
 - **Asks for details.** How often an answer asks for more information. This is measured on the 7 questions where the expert also asked, and on the other 253, where asking less is better.
+- **Reference answer.** For each question, the judge also scored an answer from Claude Sonnet 5.5 with the same search. It isn't in the table, which compares models of the same size.
 </details>