Instructions to use burtenshaw/openenv-echo-world-model-distilgpt2-seed1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use burtenshaw/openenv-echo-world-model-distilgpt2-seed1 with Transformers:

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

pipe = pipeline("text-generation", model="burtenshaw/openenv-echo-world-model-distilgpt2-seed1")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("burtenshaw/openenv-echo-world-model-distilgpt2-seed1")
model = AutoModelForCausalLM.from_pretrained("burtenshaw/openenv-echo-world-model-distilgpt2-seed1")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use burtenshaw/openenv-echo-world-model-distilgpt2-seed1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "burtenshaw/openenv-echo-world-model-distilgpt2-seed1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "burtenshaw/openenv-echo-world-model-distilgpt2-seed1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/burtenshaw/openenv-echo-world-model-distilgpt2-seed1

SGLang

How to use burtenshaw/openenv-echo-world-model-distilgpt2-seed1 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 "burtenshaw/openenv-echo-world-model-distilgpt2-seed1" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "burtenshaw/openenv-echo-world-model-distilgpt2-seed1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "burtenshaw/openenv-echo-world-model-distilgpt2-seed1" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "burtenshaw/openenv-echo-world-model-distilgpt2-seed1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use burtenshaw/openenv-echo-world-model-distilgpt2-seed1 with Docker Model Runner:
```
docker model run hf.co/burtenshaw/openenv-echo-world-model-distilgpt2-seed1
```

OpenEnv ECHO World Model

This checkpoint was trained with examples/echo_world_model/train_echo.py to predict OpenEnv terminal environment outputs from verifier-free ECHO loss.

Training Metrics

metric	value
`best_step`	`45`
`heldout_ce_after`	`0.26160934567451477`
`heldout_ce_before`	`6.333791732788086`
`heldout_ce_delta`	`-6.072182387113571`
`heldout_ce_improvement_pct`	`95.8696250727626`
`heldout_token_acc_after`	`0.8947368421052632`
`heldout_token_acc_before`	`0.10526315789473684`
`lr`	`5e-05`
`model`	`distilgpt2`
`seed`	`1`
`steps`	`60`

Downloads last month: 18

Safetensors

Model size

81.9M params

Tensor type

F32

Model tree for burtenshaw/openenv-echo-world-model-distilgpt2-seed1

Base model

distilbert/distilgpt2

Finetuned

(1520)

this model