Compare / Qwen3-235B-A22B-Thinking-2507 vs DeepSeek-V3.2

Qwen3-235B-A22B-Thinking-2507 vs DeepSeek-V3.2

Which is cheaper, which is more capable, and which does better at each kind of work, from 18 results on 3 sources that measured both models.

AlibabaQwen3-235B-A22B-Thinking-2507 Released 25 Jul 2025 · $0.30 in · $3.00 out
DeepSeekDeepSeek-V3.2 Released 1 Dec 2025 · $0.30 in · $0.96 out
Qwen3-235B-A22B-Thinking-2507 wins 02 too close to callDeepSeek-V3.2 wins 16

Of the 18 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.

Results as of 9 October 2026 (catalogue release 2026-10-09-dd620e778ac5). See also Alibaba vs DeepSeek, every model from both labs.

Qwen3-235B-A22B-Thinking-2507 or DeepSeek-V3.2? The short answer

Pick DeepSeek-V3.2: it scores higher (Intelligence Index 21.5 against 12.7) and costs less.

  • DeepSeek-V3.2 is cheaper: $0.46 against $0.97 per million tokens, 2.1 times less.
  • DeepSeek-V3.2 scores higher on the Artificial Analysis Intelligence Index: 21.5 against 12.7.
  • DeepSeek-V3.2 is newer: released 1 Dec 2025, against 25 Jul 2025 for Qwen3-235B-A22B-Thinking-2507.
  • On third-party benchmarks, on the results both models have: DeepSeek-V3.2 does better for agents and tool use, reasoning and knowledge and writing.

Head to head

12.7
Intelligence Index
21.5
$0.97
Price per million tokens, blendedlower is better
$0.46
131,072
Context window, tokens
163,840

Every fact side by side

Qwen3-235B-A22B-Thinking-2507DeepSeek-V3.2
Price per million tokens$0.30 in · $3.00 out$0.30 in · $0.96 out
Blended (3 in : 1 out)$0.97$0.46
Cheapest host, blended$0.75 (Alibaba)$0.23 (GMICloud, FP8)
Intelligence Index12.721.5
Spring Prompt overall––
Output speed, tokens a second––
Context window131,072 tokens163,840 tokens
Released25 Jul 20251 Dec 2025 (newer)
WeightsOpenOpen
Developer based inChinaChina

Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.

Which is better for what

Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.

Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 138 of 164 87 of 164
Reasoning and knowledgeHard questions across science, maths and general knowledge 117 of 198 85 of 198
WritingWhat people prefer in blind comparisons, and judged writing quality 122 of 188 65 of 188
Sticking to the factsSummarising without inventing things, and factual answers 92 of 152 58 of 152

Every shared result

Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.

MeasureQwen3-235B-A22B-Thinking-2507DeepSeek-V3.2
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Overall · TextArena rating, higher is better 139 · 106 of 387 measured 1,400 1,425
Business, management and finance · TextArena rating, higher is better 133 · 102 of 380 measured 1,404 1,423thinking reasoning
Creative writing · TextArena rating, higher is better 133 · 93 of 385 measured 1,371 1,401
Artificial Analysis reported by Artificial Analysis · 3 results
Artificial Analysis Intelligence Indexindex score, higher is better 147 · 96 of 243 measured 12.7 21.5thinking reasoning
GPQA Diamond% of questions, higher is better 114 · 89 of 231 measured 79.0% 84.0%thinking reasoning
Humanity's Last Exam% of questions, higher is better 122 · 93 of 242 measured 15.9% 24.6%thinking reasoning
UGI Leaderboard reported by UGI Leaderboard · 3 results
Writing score · UGIscore out of 100, higher is better 139 · 54 of 230 measured 32.4 57.3thinking reasoning
Requested-length error · UGI% off the requested word count, lower is better 166 · 151 of 230 measured 30.0% 26.0%no reasoning
Style adherence · UGIscore from 0 to 1, higher is better 194 · 55 of 230 measured 0.31 0.37thinking reasoning
Show all 18 results (9 more)
MeasureQwen3-235B-A22B-Thinking-2507DeepSeek-V3.2
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Expert prompts · TextArena rating, higher is better 85 · 94 of 338 measured 1,466 1,452thinking reasoning
Instruction following · TextArena rating, higher is better 142 · 94 of 387 measured 1,385 1,420
Writing, literature and language · TextArena rating, higher is better 139 · 91 of 386 measured 1,377 1,410
Artificial Analysis reported by Artificial Analysis · 6 results
Terminal-Bench 2.1% of tasks, higher is better 101 · 65 of 130 measured 12.0% 46.8%thinking reasoning
Artificial Analysis Coding Indexindex score, higher is better 98 · 67 of 131 measured 22.1 44.2thinking reasoning
IFBench% of instructions, higher is better 101 · 79 of 195 measured 51.2% 60.7%thinking reasoning
Long-context reasoning (AA-LCR)% of questions, higher is better 99 · 92 of 233 measured 72.0% 73.3%thinking reasoning
Terminal-Bench Hard% of tasks, higher is better 116 · 47 of 190 measured 13.6% 35.6%thinking reasoning
τ²-bench telecom% of tasks, higher is better 101 · 40 of 193 measured 53.2% 90.6%thinking reasoning

Quick answers

Is Qwen3-235B-A22B-Thinking-2507 better than DeepSeek-V3.2?

Pick DeepSeek-V3.2: it scores higher (Intelligence Index 21.5 against 12.7) and costs less.

Which is cheaper, Qwen3-235B-A22B-Thinking-2507 or DeepSeek-V3.2?

DeepSeek-V3.2 is cheaper: $0.46 against $0.97 per million tokens, 2.1 times less.

Which is better for agents and tool use, Qwen3-235B-A22B-Thinking-2507 or DeepSeek-V3.2?

DeepSeek-V3.2 ranks 87 of 164 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 138 for Qwen3-235B-A22B-Thinking-2507.

Which is better for reasoning and knowledge, Qwen3-235B-A22B-Thinking-2507 or DeepSeek-V3.2?

DeepSeek-V3.2 ranks 85 of 198 models on sale for hard questions across science, maths and general knowledge, against 117 for Qwen3-235B-A22B-Thinking-2507.

Which is better for writing, Qwen3-235B-A22B-Thinking-2507 or DeepSeek-V3.2?

DeepSeek-V3.2 ranks 65 of 188 models on sale for what people prefer in blind comparisons, and judged writing quality, against 122 for Qwen3-235B-A22B-Thinking-2507.

Which is better for sticking to the facts, Qwen3-235B-A22B-Thinking-2507 or DeepSeek-V3.2?

DeepSeek-V3.2 ranks 58 of 152 models on sale for summarising without inventing things, and factual answers, against 92 for Qwen3-235B-A22B-Thinking-2507.

Which has the bigger context window, Qwen3-235B-A22B-Thinking-2507 or DeepSeek-V3.2?

DeepSeek-V3.2: 163,840 tokens, against 131,072 for Qwen3-235B-A22B-Thinking-2507.

Run Qwen3-235B-A22B-Thinking-2507 and DeepSeek-V3.2 on your own prompt

Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.

Try the playgroundTalk to us