DeepSeek R1 Distill Qwen 1.5B

Overview

DeepSeek's R1 reasoning distilled into a 1.5B MIT-licensed model with visible chain-of-thought. Hits MATH-500 83.9 and runs on any laptop.

When to pick this model

• Teaching and demos showing CoT reasoning on minimal hardware
• Math tutoring apps on edge devices
• Research baselines for distillation experiments
• Battery-constrained mobile deployments

VRAM requirements by quantization

Quantization	VRAM required
Q4_K_M (recommended)	1 GB
Q5_K_M	1.2 GB
Q8_0	2 GB
FP16 (no quantization)	3 GB

VRAM figures include model weights plus a typical 8k KV cache and ~600 MB runtime overhead (Ollama / llama.cpp baseline). Add headroom for higher context lengths.

Published benchmark scores

Benchmark	Score
MATH-500	83.9

Scores published by the model author or aggregated from public leaderboards. Re-measured monthly by our editorial team.

Strengths

• Around 1GB VRAM at Q4 — runs on any laptop
• Visible chain-of-thought reasoning
• MIT license — fully unrestricted
• 128k context in a 1.5B model

Limitations

• Reasoning depth is genuinely limited at 1.5B despite CoT
• Highly verbose — token costs add up fast
• Outclassed by the 14B distill on anything non-trivial

Architecture & training

Architecture: DeepSeek R1 distillation into Qwen 2.5 1.5B · chain-of-thought

Training: Distilled from R1 671B. Ultra-compact 1.5B version with CoT reasoning.

Quick start

ollama run deepseek-r1:1.5b

Or use the open-source MCP server to query this model from Claude Desktop, Cursor, or any MCP-compatible client.