orcarouter/Qwen3.8-27B-Uncensored-FP8

API vía OrcaRouter qwen/qwen3.8-27b - 262K context, FP8, abliterated. HF | OrcaRouter

Thinking

Configura ORCA_API_KEY en el servidor: export ORCA_API_KEY=sk-orca-... o edita api/qwen.php:10. Requiere aceptar disclaimer en HF.

cURL directo OrcaRouter

curl -X POST https://api.orcarouter.ai/v1/chat/completions \
 -H "Authorization: Bearer sk-orca-..." \
 -H "Content-Type: application/json" \
 -d '{"model":"qwen/qwen3.8-27b","messages":[{"role":"user","content":"Hola"}]}'

Python (OpenAI SDK)

from openai import OpenAI
client = OpenAI(base_url="https://api.orcarouter.ai/v1", api_key="sk-orca-...")
r = client.chat.completions.create(model="qwen/qwen3.8-27b", messages=[{"role":"user","content":"Hola"}])
print(r.choices[0].message.content)

vLLM self-host (requiere 40GB VRAM)

pip install vllm
vllm serve "orcarouter/Qwen3.8-27B-Uncensored-FP8" --served-model-name Qwen3.8-27B-Uncensored
curl http://localhost:8000/v1/chat/completions -H "Content-Type: application/json" -d '{"model":"Qwen3.8-27B-Uncensored","messages":[{"role":"user","content":"Hola"}]}'