About
DeepSeek R1 Distill Qwen 32B is a distilled large language model based on [Qwen 2.5 32B](https://huggingface.co/Qwen/Qwen2.5-32B), using outputs from [DeepSeek R1](/deepseek/deepseek-r1). It outperforms OpenAI's o1-mini across various benchmarks, achieving new...
Related Models
DeepSeek: DeepSeek V4 Pro2026-04-24DeepSeek V4 Pro (Reasoning, High Effort)2026-04-24DeepSeek: DeepSeek V4 Flash2026-04-24DeepSeek V4 Flash (Reasoning, High Effort)2026-04-24DeepSeek V4 Pro (Non-reasoning)2026-04-24DeepSeek V4 Flash (Non-reasoning)2026-04-24DeepSeek: DeepSeek V4 Flash (free)2026-04-24DeepSeek V3.2 (Reasoning)2025-12-01
Pricing
Input
$0.29
per 1M tokens
Output
$0.29
per 1M tokens
Blended
$0.29
per 1M tokens
Cheaper than 63% of models. Median price is $0.54/1M tokens.
Cost Calculator
Tokens per day1M
100K100M
Daily
$0.29
Monthly
$8.70
vs. Similar Models
Z.ai: GLM 4.6VQ:0.0
$0.45+55%
Qwen: Qwen3 VL 32B InstructQ:+0.1
$0.18-37%
Ministral 3 14BQ:+0.1
$0.20-31%
Qwen3 235B A22B (Non-reasoning)Q:-0.1
$0.79+171%
Performance
43
tokens/sec
Faster than 9% of models
0.45
seconds
Faster than 91% of models
47.11
seconds
Faster than 7% of models
Market Median
94 tok/s
55% slower
Median TTFT
1.11s
60% faster
Throughput/Dollar
148
tok/s per $/1M
Speed Comparison
QwQ 32B-Preview
43 tok/s+1%
Claude Opus 4.7 (Non-reasoning, High Effort)
42 tok/s-2%
OpenAI: GPT-4
42 tok/s-2%
Context Window
128K
tokens
Larger than 16% of models
Max Output
33K
tokens
26% of context
Benchmarks
MMLU-Pro
73.9%
GPQA Diamond
61.5%
HLE
5.5%
LiveCodeBench
27.0%
SciCode
37.6%
TerminalBench HardNot evaluated
MATH-500
94.1%
AIME
68.7%
AIME 2025
63.0%
IFBench
22.9%
Long Context Recall
9.7%
Tau2Not evaluated
Market AverageTop Score
Open Source
mit32BGGUF / GPTQ / AWQ
Downloads
843.8K
Likes
1.6K
VRAM (FP16)
24-48 GB
GPU
A6000 / M3 Ultra