Instructions to use EpistemeAI/VibeCoder-20b-RL1_0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use EpistemeAI/VibeCoder-20b-RL1_0 with Transformers:

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

pipe = pipeline("text-generation", model="EpistemeAI/VibeCoder-20b-RL1_0")
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("EpistemeAI/VibeCoder-20b-RL1_0")
model = AutoModelForCausalLM.from_pretrained("EpistemeAI/VibeCoder-20b-RL1_0", 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 EpistemeAI/VibeCoder-20b-RL1_0 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "EpistemeAI/VibeCoder-20b-RL1_0"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "EpistemeAI/VibeCoder-20b-RL1_0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/EpistemeAI/VibeCoder-20b-RL1_0

SGLang

How to use EpistemeAI/VibeCoder-20b-RL1_0 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 "EpistemeAI/VibeCoder-20b-RL1_0" \
    --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": "EpistemeAI/VibeCoder-20b-RL1_0",
		"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 "EpistemeAI/VibeCoder-20b-RL1_0" \
        --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": "EpistemeAI/VibeCoder-20b-RL1_0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Desktop
Docker Model Runner
How to use EpistemeAI/VibeCoder-20b-RL1_0 with Docker Model Runner:
```
docker model run hf.co/EpistemeAI/VibeCoder-20b-RL1_0
```

legolasyiu commited on Oct 8, 2025

Commit

d242dbb

verified ·

1 Parent(s): 15732dd

Update README.md

Browse files

Files changed (1) hide show

README.md +18 -7

README.md CHANGED Viewed

@@ -60,13 +60,24 @@ Debugged vibecoder dataset
 ## Benchmark
-|      Tasks      | Version |      Filter      | n-shot |    Metric    | VibeCoder-20b-0.02-Debugger | gpt-oss-20 | Qwen 3 235B |
-|-----------------|---------|------------------|--------|--------------|-----------------------------|------------|-------------|
-| gsm8k_cot       |    3    | flexible-extract |   3    | exact_match ↑ | 0.8452+(0.7667)            |   0.78     |    0.82     |
-| humaneval       |    1    | create_test      |   0    | exact_match ↑ | 0.933+( 0.8)               |   0.73     |    0.92     |
-| mmlu_college_biology|    1    | create_test      |   0    | exact_match ↑ | 1.0                    |            |             |
-| mmlu_high_school_computer_science|    1    | create_test      |   0   | exact_match ↑ | 1.0+(0.9) |            |             |
-|computer_security|    1    |  none            |     2  |    acc ↑      |0.8528+(0.700)| |             |
 ## Example Usage

 ## Benchmark
+### 📊 Model Evaluation Results
+| Tasks                              | Version | Filter           | n-shot | Metric         | VibeCoder-20b-0.02-Debugger | gpt-oss-20 | Qwen 3 235B |
+|-----------------------------------|----------|------------------|--------|----------------|-----------------------------|-------------|--------------|
+| gsm8k_cot                         | 3        | flexible-extract | 3      | exact_match ↑  | 0.8452 (+0.7667)           | 0.78        | 0.82         |
+| humaneval                         | 1        | create_test      | 0      | exact_match ↑  | 0.933 (+0.8)               | 0.73        | 0.92         |
+| mmlu_college_biology              | 1        | create_test      | 0      | exact_match ↑  | 1.000 (+ – )               | —           | —            |
+| mmlu_high_school_computer_science | 1        | create_test      | 0      | exact_match ↑  | 1.000 (+0.9)               | —           | —            |
+| computer_security                 | 1        | none             | 2      | acc ↑          | 0.8528 (+0.700)            | —           | —            |
+| college_computer_science          | 1        | none             | 2      | acc ↑          | 0.8528 (+0.700)            | —           | —            |
+---
+**Notes:**
+- The `(+value)` indicates delta over baseline evaluation.
+- Metrics marked with `↑` denote that higher is better.
+- Dashes (`—`) indicate results not yet reported or evaluated.
 ## Example Usage