DeepSeek R1-0528 Qwen3 8B

Reasoning model — R1-0528 distilled onto Qwen3 8B.

Overview

• model_name: hf-deepseek-ai-DeepSeek-R1-0528-Qwen3-8B (HuggingFace import, currently free)
• Source: deepseek-ai/DeepSeek-R1-0528-Qwen3-8B
• Scale: ~8B parameters
• Served on: GPU via vLLM — verified on both NVIDIA (CUDA) and Intel Arc (XPU) nodes
• Status: ready (verified end-to-end through the public inference route, 2026-08-02)

When to use

✅ Chain-of-thought reasoning, math and logic problems, and tasks that benefit from the R1 "thinking" style at a small-model cost.

❌ For plain fast chat without reasoning overhead, prefer Qwen 2.5 7B — R1-style models spend tokens thinking before answering.

Live specs

Context window, VRAM footprint, and readiness come from the live catalog:

curl "https://api.colabhive.com/api/builder/v1/endpoints?visibility=public&search=DeepSeek-R1-0528"

Quick start

curl -X POST "https://api.colabhive.com/api/builder/v1/endpoints/<endpoint_id>/infer" \
  -H "X-API-Key: $COLABHIVE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"input": {"messages": [{"role": "user", "content": "Prove that sqrt(2) is irrational."}], "max_tokens": 512}}'

The response streams the model's reasoning followed by the final answer.