DeepSeek-R1-Distill-Qwen-32B: benchmark scores, price and ranking

DeepSeek-R1-Distill-Qwen-32B from DeepSeek ranks #395 of 743 LLMs on the LLMs Tiger aggregate benchmark score (47.9), based on 5 benchmark results captured up to 2026-10-13.

Organisation: DeepSeek
Type: Open-weight / local model
Open weights: Yes
Released: 2025-01-20 (source: epoch)
Cheapest API price: $0.150 in / $0.150 out per million tokens (via nscale, observed 2026-10-13T20:47Z)
Aggregate rank: #395 of 743 (score 47.9)

Relative to other models, DeepSeek-R1-Distill-Qwen-32B is strongest on Epoch ECI (#62 of 204) and weakest on Chess Puzzles (#94 of 122).

Across the 5 benchmarks where at least ten models are scored, DeepSeek-R1-Distill-Qwen-32B is in the top 10% on 0 and in the top half on 1.

Its cheapest listed input price, $0.150 per million tokens, is lower than that of 72% of the 427 models with a listed price.

DeepSeek-R1-Distill-Qwen-32B benchmark results

DeepSeek-R1-Distill-Qwen-32B benchmark scores, ranked against all models measured on each benchmark
Benchmark	Score	Rank	Source	Captured
Epoch ECI	137	#62 of 204	Epoch AI Benchmarking Hub	2026-10-13
OTIS Mock AIME	55.6%	#99 of 189	Epoch AI Benchmarking Hub	2026-10-13
GPQA Diamond	64.1%	#160 of 275	Epoch AI Benchmarking Hub	2026-10-13
BALROG	19.5%	#21 of 34	Epoch AI Benchmarking Hub	2026-10-13
Chess Puzzles	1.0%	#94 of 122	Epoch AI Benchmarking Hub	2026-10-13

About DeepSeek-R1-Distill-Qwen-32B

DeepSeek-R1-Distill-Qwen-32B is an open-weight model that can run on your own hardware from DeepSeek. It was released on 2025-01-20, according to Epoch AI. The cheapest API listing we track charges $0.150 per million input tokens and $0.150 per million output tokens, through nscale. We hold 5 benchmark results for it across 3 areas, and none is in the top tenth of its board.

Where DeepSeek-R1-Distill-Qwen-32B stands

On our aggregate score, DeepSeek-R1-Distill-Qwen-32B is #395 of 743 ranked models, around the middle of the table. Its score is 47.9; the scale is centred on 50, so it sits 2.1 points below the typical model. That beats 47% of the ranked field. Among the 33 ranked DeepSeek models it is #20. The next one up is DeepSeek-R1 at 48.0. The next one down is DeepSeek-V3.2 (Thinking; Fireworks) at 47.9.

DeepSeek-R1-Distill-Qwen-32B, benchmark by benchmark

Epoch ECI puts DeepSeek-R1-Distill-Qwen-32B at 137, #62 of 204. That beats 70% of the models tested, but 19 points short of Qwen 3.8 Max (156). The results either side are QwQ-32B (138) and Qwen3-30B-A3B-Instruct (Jul 2025) (137). It covers Epoch Capabilities Index: Epoch's own aggregate of ~40 benchmarks. The board median is 127, so DeepSeek-R1-Distill-Qwen-32B is 10 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

DeepSeek-R1-Distill-Qwen-32B lands below the middle on OTIS Mock AIME: 55.6%, #99 of 189, with 52% of the field scoring higher. Claude Fable 5.1 (max) leads at 100.0%; the closest result above is qwen3.5-4b (none) at 55.8%. It covers competition maths (AIME 2024-25 style). The board median is 57.8%, so DeepSeek-R1-Distill-Qwen-32B is 2.2 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

DeepSeek-R1-Distill-Qwen-32B lands below the middle on GPQA Diamond: 64.1%, #160 of 275, with 58% of the field scoring higher. GPT-6 Astra (max) leads at 95.8%; the closest result above is QwQ-32B at 65.3%. It covers expert-level science questions. The board median is 71.7%, so DeepSeek-R1-Distill-Qwen-32B is 7.6 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

DeepSeek-R1-Distill-Qwen-32B lands below the middle on BALROG: 19.5%, #21 of 34, with 59% of the field scoring higher. GPT-6 Astra (max) leads at 68.3%; the closest result above is gemini-1.5-pro-002 at 21.0%. It covers long-horizon game-playing agents. The board median is 27.6%, so DeepSeek-R1-Distill-Qwen-32B is 8.1 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

DeepSeek-R1-Distill-Qwen-32B lands below the middle on Chess Puzzles: 1.0%, #94 of 122, with 76% of the field scoring higher. GPT-6 Astra (max) leads at 72.0%; the closest result above is DeepSeek-R1-Distill-Qwen-14B at 1.0%. It covers chess tactics puzzles. The board median is 12.5%, so DeepSeek-R1-Distill-Qwen-32B is 11.5 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

Strengths and weaknesses

Reasoning, maths and science: uneven, ahead of 45% of the field on average across 2 benchmarks. Its best board here is OTIS Mock AIME (#99 of 189) and its weakest is GPQA Diamond (#160 of 275). General knowledge and chat: uneven, ahead of 46% of the field on average across 2 benchmarks. Its best board here is Epoch ECI (#62 of 204) and its weakest is Chess Puzzles (#94 of 122).

Price and value

The median input price across the 427 priced models is $0.500 per million tokens, so DeepSeek-R1-Distill-Qwen-32B costs less than the typical listing. Models that score within 1.5 points or better and cost less: Qwen2.5-Coder-7B-Instruct ($0.010, 49.5), DeepSeek V4.1 Flash (high) ($0.010, 48.9), Qwen2.5-Coder-3B-Instruct ($0.010, 46.5). Similar scores cost more on Phi-3-medium-4k-instruct ($0.170) and qwen3-235b-a22b ($0.180).

Release history

The DeepSeek model released before it was DeepSeek-R1-Distill-Qwen-14B on 2025-01-20, which scores 47.0. The next one was DeepSeek V3 0324 on 2025-03-24, at 50.8.

Reading these numbers

The 5 scores on this page come from 1 source: Epoch AI Benchmarking Hub (5). No official leaderboard supplied any of them, so treat them as third-party measurements. With 5 results the aggregate rank can move a lot when one new score arrives.

Models ranked nearby

#392 Codestral-22B-v0.1-Q4_K_M (48.0)
#393 codestral:22b-v0.1-q8_0 (48.0)
#394 qwen3-4b (47.9)
#396 DeepSeek-V3.2 (Thinking; Fireworks) (47.9)
#397 qwen3-235b-a22b (47.9)
#398 Opus 4.5 (Thinking, None) (47.9)