Instructions to use mis2/Trajectory2Gym-4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mis2/Trajectory2Gym-4B with Transformers:

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

pipe = pipeline("text-generation", model="mis2/Trajectory2Gym-4B")
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("mis2/Trajectory2Gym-4B")
model = AutoModelForCausalLM.from_pretrained("mis2/Trajectory2Gym-4B", 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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use mis2/Trajectory2Gym-4B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mis2/Trajectory2Gym-4B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mis2/Trajectory2Gym-4B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mis2/Trajectory2Gym-4B

SGLang

How to use mis2/Trajectory2Gym-4B 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 "mis2/Trajectory2Gym-4B" \
    --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": "mis2/Trajectory2Gym-4B",
		"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 "mis2/Trajectory2Gym-4B" \
        --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": "mis2/Trajectory2Gym-4B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use mis2/Trajectory2Gym-4B with Docker Model Runner:
```
docker model run hf.co/mis2/Trajectory2Gym-4B
```

nielsr HF Staff commited on 17 days ago

Commit

5a79cfa

verified ·

1 Parent(s): d182266

Add links to paper, project page and code

Browse files

This PR adds links to the paper ([Trajectory2Gym: Decision-Grounded Compositional Interaction Synthesis for Tool-Use Agents](https://huggingface.co/papers/2610.33568)), the [project page](https://yihaohu0118.github.io/Trajectory2Gym-page/), and the [GitHub repository](https://github.com/yihaohu0118/Trajectory2Gym) to the model card, making it easier for users to find additional information about the model and its implementation.

Files changed (1) hide show

README.md +11 -6

README.md CHANGED Viewed

@@ -1,21 +1,26 @@
 ---
-license: apache-2.0
 library_name: transformers
 pipeline_tag: text-generation
-base_model: Qwen/Qwen3-4B
-base_model_relation: finetune
 tags:
 - trajectory2gym
 - tool-use
 - supervised-fine-tuning
-datasets:
-- mis2/Trajectory2gym
 ---
 # Trajectory2Gym-4B
 Full-parameter supervised post-training for decision-grounded tool use. This checkpoint is the paper's corresponding scale, not a support-depth or component-ablation model.
 ## Reported evaluation
 | Tau1 | Tau2 | BFCL three-group Overall |
@@ -43,4 +48,4 @@ The public candidate corpus is not the exact rendered training view. The 30B-A3B
 Weight files total 17,645,742,616 bytes. The source training-view SHA-256 is ab2e580badcec821f5cd3a8937d41366b75220ea6edcf2e9835d1a3e5a0de8cf. RELEASE_MANIFEST.json provides per-file checksums and archive-byte verification details. Internal source paths, credentials, optimizer states, training arguments and raw logs are excluded. Only non-inference path metadata may be removed from JSON configurations, as recorded in the manifest.
-Apache-2.0; the base model's license is retained as LICENSE. Tool execution should use permissioned environments and independent authorization for high-impact actions.

 ---
+base_model: Qwen/Qwen3-4B
+datasets:
+- mis2/Trajectory2gym
 library_name: transformers
+license: apache-2.0
 pipeline_tag: text-generation
 tags:
 - trajectory2gym
 - tool-use
 - supervised-fine-tuning
+base_model_relation: finetune
 ---
 # Trajectory2Gym-4B
 Full-parameter supervised post-training for decision-grounded tool use. This checkpoint is the paper's corresponding scale, not a support-depth or component-ablation model.
+This model is presented in the paper [Trajectory2Gym: Decision-Grounded Compositional Interaction Synthesis for Tool-Use Agents](https://huggingface.co/papers/2610.33568).
+Project page: https://yihaohu0118.github.io/Trajectory2Gym-page/
+Code: https://github.com/yihaohu0118/Trajectory2Gym
 ## Reported evaluation
 | Tau1 | Tau2 | BFCL three-group Overall |
 Weight files total 17,645,742,616 bytes. The source training-view SHA-256 is ab2e580badcec821f5cd3a8937d41366b75220ea6edcf2e9835d1a3e5a0de8cf. RELEASE_MANIFEST.json provides per-file checksums and archive-byte verification details. Internal source paths, credentials, optimizer states, training arguments and raw logs are excluded. Only non-inference path metadata may be removed from JSON configurations, as recorded in the manifest.
+Apache-2.0; the base model's license is retained as LICENSE. Tool execution should use permissioned environments and independent authorization for high-impact actions.