[whatmodel.sh](../../README.md) / DeepSeek V3.2 Exp
# DeepSeek V3.2 Exp
DeepSeek · released 2025-09 · 671B params (37B active) · mit
**score 65.9** ±0.9 · #128 of 732 · from 14 benchmark results across 6 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks DeepSeek V3.2 Exp among every model on that task or test, worst to best.
task vs the others rank counts tests
agent tasks ▁▁▃▆█▄●▅▄▄▂▁ #151 of 313* 4.1% 1
coding ▁▁▁▃▅▇█●▅▄▃▁ #144 of 469 43% 4
chat & writing ▁▁▁▃▆▄▆█▆●▇▃ #86 of 402* 8.6% 1
math ▁▁▃▅█▆▄●▄▂▁▁ #75 of 431 14% 3
agent tasks · #151 of 313 · 4.1% of its score
Trails [SmolLM3 3B](smollm3-3b.md), [Jamba 1.5 Large](jamba-1-5-large.md), [GPT-4o (2024-05-13)](gpt-4o-2024-05-13.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
BFCL V4 (December 2025 archive) 54.1% ▁▁▁▂██▅▅●▂▂▂ beats 63/77 hard · avg 31% 4.1% at its level
coding · #144 of 469 · 43% of its score
Beats [Qwen3.5 Plus](qwen3-5-plus.md), [Qwen3-Coder-Next (80B-A3B)](qwen3-coder-next.md), [Gemini 2.5 Pro](gemini-2-5-pro.md) here, though they rank higher overall. Trails [Llama 3.1 Nemotron Ultra 253B v1](llama-3-1-nemotron-ultra-253b.md), [Devstral Small 2 24B](devstral-small-2.md), [Gemma 4 12B](gemma-4-12b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
Aider polyglot (225 exercises, tw… 74.2% ▂·▂▃▅▂▃█▆●▂▂ beats 43/49 very hard · avg 7.6% 21% sharp · at its level
Arena WebDev (Code Arena) 1271.4 ▁▁▃●▅█▃▆▅▅▁▂ beats 17/129 avg 981 15% sharp
WebApp1K-Duo-React pass@1 (onekq-… 64.1% ▁··▁·▁▁▁▆●▇█ beats 19/45 hard · avg 29% 4.5% at its level
LiveCodeBench v6 (Aug 2024–May 20… 71.6% ▄▄·▄·····█·● beats 5/5 medium · avg 45% 2.8% at its level
chat & writing · #86 of 402 · 8.6% of its score
Beats [Claude Haiku 5.5](claude-haiku-5-5.md), [LongCat 2.0 (1.6T-A48B)](longcat-2-0.md), [Inkling Small (266B-A12B)](inkling-small.md) here, though they rank higher overall. Trails [Gemma 4 12B](gemma-4-12b.md), [Gemma 4 26B-A4B](gemma-4-26b-a4b.md), [Mistral Medium 3.5 (128B)](mistral-medium-3-5.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
LMArena (text) 1425.1 ▁▁▁▂▃▃▃▆▆●█▃ beats 280/385 avg 1260 8.6% blunt
math · #75 of 431 · 14% of its score
Beats [Claude Sonnet 4.6](claude-sonnet-4-6.md), [Claude Opus 4.5](claude-opus-4-5.md), [Qwen3.7 Plus](qwen3-7-plus.md) here, though they rank higher overall. Trails [Granite 4.2 30B](granite-4-2-30b.md), [INTELLECT-3](intellect-3.md), [Qwen3 235B-A22B Thinking 2507](qwen3-235b-a22b-thinking-2507.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
AIME 2024 88.1% █···███·●··█ beats 4/5 medium · avg 50% 7.2% sharp
AIME 2025 84.7% █·█···██●··█ beats 4/5 medium · avg 42% 6.5% sharp
MATH-500 96.8% ▄······▄●▄·█ beats 2/5 very easy · avg 86% 0.7% maxed out · noisy
other tests (5)
test result vs the others in its run difficulty counts why
GPQA Diamond 81.4% █··██·██···● beats 5/5 hard · avg 41% 16% sharp · at its level
Humanity's Last Exam 17.9% ▄····▄·█·▄·● beats 5/5 very hard · avg 4.1% 9.1% typical
MMLU-Pro 84.6% ▁▁·▁▁▁▃▂▄▅●▁ beats 166/188 medium · avg 50% 4.3% typical
Vectara hallucination leaderboard… 95.4% ▁▁▂▃▄▄●▆▃▂▂▁ beats 73/140 very easy · avg 94% 0.3% blunt · maxed out · noisy
Vectara hallucination leaderboard… 94.7% ▁▂▁▂▅█▂▃▄●▁▁ beats 86/98 very easy · avg 90% 0.0% blunt
source results share of the score
Prime Intellect INTELLECT-3 model card:… 7 46% ▇▇▇▇▇▇
LMArena leaderboard dataset 2 24% ▇▇▇
Aider polyglot benchmark runs 1 21% ▇▇
WebApp1K models leaderboard 1 4% ▇
Berkeley Function Calling Leaderboard r… 1 4% ▇
Vectara hallucination leaderboard 2 0% ▇
prior: pull toward the middle 0% ▇
weighted average = 65.9
Licences: Prime Intellect INTELLECT-3 model card: Evaluation table ([MIT](https://huggingface.co/PrimeIntellect/INTELLECT-3/blob/ff39d4a4688989f3f28868923d030c28e1b7d81c/README.md)) · LMArena leaderboard dataset ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Aider polyglot benchmark runs ([Apache-2.0](../../licenses/aider-polyglot-runs-Apache-2.0.txt)) · WebApp1K models leaderboard ([MIT](../../licenses/webapp1k-leaderboard-MIT.txt)) · Berkeley Function Calling Leaderboard result archive ([Apache-2.0](../../licenses/bfcl-result-archive-v4-Apache-2.0.txt)) · Vectara hallucination leaderboard ([Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0))
## How to run it
### Other ways to run it
1. Run it as published (fp8, needs ~689.48 GB, NVIDIA only, from [huggingface.co/deepseek-ai](https://huggingface.co/deepseek-ai/DeepSeek-V3.2-Exp/blob/194c67e12b1b0d6df0ef373ddcf215bc84027409/README.md))
```sh
python -m sglang.launch_server --model deepseek-ai/DeepSeek-V3.2-Exp --tp 8 --dp 8 --enable-dp-attention
```
Other versions: fp8 (~689.48 GB, NVIDIA only) `vllm serve deepseek-ai/DeepSeek-V3.2-Exp`
Model repository and documentation: [huggingface.co/deepseek-ai](https://huggingface.co/deepseek-ai/DeepSeek-V3.2-Exp)
[interactive version](https://whatmodel.sh/model/deepseek-v3-2-exp) · [JSON](https://whatmodel.sh/api?format=json&all=1)