Instructions to use CodeXomics/CodeXomics-ToolAgent-4B-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use CodeXomics/CodeXomics-ToolAgent-4B-v1 with Transformers:

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

pipe = pipeline("text-generation", model="CodeXomics/CodeXomics-ToolAgent-4B-v1")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("CodeXomics/CodeXomics-ToolAgent-4B-v1")
model = AutoModelForMultimodalLM.from_pretrained("CodeXomics/CodeXomics-ToolAgent-4B-v1", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use CodeXomics/CodeXomics-ToolAgent-4B-v1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "CodeXomics/CodeXomics-ToolAgent-4B-v1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CodeXomics/CodeXomics-ToolAgent-4B-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/CodeXomics/CodeXomics-ToolAgent-4B-v1

SGLang

How to use CodeXomics/CodeXomics-ToolAgent-4B-v1 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 "CodeXomics/CodeXomics-ToolAgent-4B-v1" \
    --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": "CodeXomics/CodeXomics-ToolAgent-4B-v1",
		"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 "CodeXomics/CodeXomics-ToolAgent-4B-v1" \
        --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": "CodeXomics/CodeXomics-ToolAgent-4B-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use CodeXomics/CodeXomics-ToolAgent-4B-v1 with Docker Model Runner:
```
docker model run hf.co/CodeXomics/CodeXomics-ToolAgent-4B-v1
```

CodeXomics commited on 3 days ago

Commit

f713245

verified ·

1 Parent(s): 19db9e9

Upload README.md with huggingface_hub

Browse files

Files changed (1) hide show

README.md +33 -10

README.md CHANGED Viewed

@@ -42,16 +42,39 @@ integration. Source code and documentation:
 ## Evaluation
 On the CodeXomics Benchmark (172 automatic tests: 143 single-operation, 29 multi-step), evaluated
-in the real application loop with task-completion scoring plus execution evidence:
-| Suite | Passed / Total |
-| --- | --- |
-| Simple | 143/143 |
-| Complex | 29/29 |
-| **Total** | **172/172 (100%)** |
-100% was achieved in multiple independent complete sessions (simple: 2026-08-04 and 2026-08-07;
-complex: 2026-08-04 and 2026-08-10). Inference settings: temperature 0, thinking enabled.
 ## Usage

 ## Evaluation
 On the CodeXomics Benchmark (172 automatic tests: 143 single-operation, 29 multi-step), evaluated
+in the real application loop with task-completion scoring plus execution evidence. Both the
+fine-tuned model and the un-fine-tuned Qwen3.5-4B baseline were evaluated in three independent
+complete sessions; results were identical across sessions for both models.
+| Suite | Qwen3.5-4B (base) | CodeXomics-ToolAgent-4B-v1 |
+| --- | ---: | ---: |
+| Simple | 139/143 | 143/143 |
+| Complex | 26/29 | 29/29 |
+| **Total** | **165/172 (95.9%)** | **172/172 (100%)** |
+Fine-tuning improved the overall accuracy by 7 tests (+4 simple, +3 complex). Inference settings:
+temperature 0, thinking enabled.
+Inference speed (mean ± SD over three independent runs; offline harness, 172 tests, Ollama Q4_K_M
+on Apple M3 Max):
+| Metric | Qwen3.5-4B (base) | CodeXomics-ToolAgent-4B-v1 |
+| --- | ---: | ---: |
+| Average latency per test (s) | 12.0 ± 0.1 | 10.7 ± 0.9 |
+| &nbsp;&nbsp;Simple suite (s) | 9.2 ± 0.1 | 8.6 ± 0.7 |
+| &nbsp;&nbsp;Complex suite (s) | 25.8 ± 0.3 | 21.5 ± 1.9 |
+| Average latency per tool call (s) | 7.9 ± 0.1 | 7.3 ± 0.6 |
+| Generation throughput (tokens/s) | 33.9 ± 0.4 | 33.9 ± 2.8 |
+| Generated tokens per test | 408 (identical across runs) | 363 (identical across runs) |
+| Prompt tokens per test | 9,004 (identical across runs) | 8,777 (identical across runs) |
+![CodeXomics Benchmark: base vs. fine-tuned 4B](codexomics-v5-benchmark-comparison.png)
+**Figure 1.** Two-panel comparison between the un-fine-tuned Qwen3.5-4B baseline and
+CodeXomics-ToolAgent-4B-v1. Left: CodeXomics Benchmark pass rates (simple/complex/total; three
+sessions each, identical results). Right: average inference latency per test with error bars
+(mean ± SD over three runs; Ollama Q4_K_M, Apple M3 Max). Latency values are the mean of the
+per-suite rows above; error bars show the run-to-run SD.
 ## Usage