Compare / Qwen3-32B vs Kimi K2 (0905)
Qwen3-32B vs Kimi K2 (0905)
Which is cheaper, which is more capable, and which does better at each kind of work, from 18 results on 4 sources that measured both models.
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 Moonshot AI, every model from both labs.
Qwen3-32B or Kimi K2 (0905)? The short answer
Pick Kimi K2 (0905) if quality matters most: it scores higher (Intelligence Index 15.3 against 8.6). Pick Qwen3-32B if cost matters more: it costs 5.2 times less per token.
- Qwen3-32B is cheaper: $0.21 against $1.07 per million tokens, 5.2 times less.
- Kimi K2 (0905) scores higher on the Artificial Analysis Intelligence Index: 15.3 against 8.6.
- Kimi K2 (0905) is newer: released 5 Sep 2025, against 28 Apr 2025 for Qwen3-32B.
- On third-party benchmarks, on the results both models have: Qwen3-32B does better for sticking to the facts; Kimi K2 (0905) for reasoning and knowledge and writing.
Head to head
Every fact side by side
| Qwen3-32B | Kimi K2 (0905) | |
|---|---|---|
| Price per million tokens | $0.14 in · $0.40 out | $0.60 in · $2.50 out |
| Blended (3 in : 1 out) | $0.21 | $1.07 |
| Cheapest host, blended | $0.13 (DeepInfra, FP8) | – |
| Intelligence Index | 8.6 | 15.3 |
| Spring Prompt overall | – | – |
| Output speed, tokens a second | – | – |
| Context window | 131,072 tokens | 262,144 tokens |
| Released | 28 Apr 2025 | 5 Sep 2025 (newer) |
| Weights | Open | Open |
| Developer based in | China | China |
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.
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.
| Measure | Qwen3-32B | Kimi K2 (0905) |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results | ||
| Overall · TextArena rating, higher is better 195 · 113 of 387 measured | 1,347 | 1,419 |
| Business, management and finance · TextArena rating, higher is better 203 · 134 of 380 measured | 1,340 | 1,404 |
| Creative writing · TextArena rating, higher is better 205 · 119 of 385 measured | 1,304 | 1,381 |
| Artificial Analysis reported by Artificial Analysis · 3 results | ||
| Artificial Analysis Intelligence Indexindex score, higher is better 188 · 126 of 243 measured | 8.6thinking reasoning | 15.3 |
| GPQA Diamond% of questions, higher is better 167 · 130 of 231 measured | 66.8%thinking reasoning | 76.7% |
| Humanity's Last Exam% of questions, higher is better 168 · 178 of 242 measured | 7.4%thinking reasoning | 6.4% |
| UGI Leaderboard reported by UGI Leaderboard · 3 results | ||
| Writing score · UGIscore out of 100, higher is better 134 · 61 of 230 measured | 33.0no reasoning | 54.4 |
| Requested-length error · UGI% off the requested word count, lower is better 99 · 129 of 230 measured | 19.0%no reasoning | 22.0% |
| Style adherence · UGIscore from 0 to 1, higher is better 175 · 194 of 230 measured | 0.32no reasoning | 0.31 |
| Vectara Hallucination Leaderboard reported by Vectara Hallucination Leaderboard · 2 results | ||
| Hallucination rate · Vectara% of summaries, lower is better 24 · 94 of 104 measured | 5.9% | 17.9% |
| Answer rate · Vectara% of documents, higher is better 17 · 72 of 104 measured | 99.9% | 98.6% |
Show all 18 results (7 more)
| Measure | Qwen3-32B | Kimi K2 (0905) |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results | ||
| Expert prompts · TextArena rating, higher is better 159 · 138 of 338 measured | 1,398 | 1,417 |
| Instruction following · TextArena rating, higher is better 198 · 136 of 387 measured | 1,331 | 1,390 |
| Writing, literature and language · TextArena rating, higher is better 217 · 133 of 386 measured | 1,308 | 1,382 |
| Artificial Analysis reported by Artificial Analysis · 4 results | ||
| IFBench% of instructions, higher is better 155 · 126 of 195 measured | 36.3%thinking reasoning | 41.7% |
| Long-context reasoning (AA-LCR)% of questions, higher is better 222 · 146 of 233 measured | 0.0%no reasoning | 53.0% |
| Terminal-Bench Hard% of tasks, higher is better 169 · 84 of 190 measured | 3.0%thinking reasoning | 23.5% |
| τ²-bench telecom% of tasks, higher is better 135 · 80 of 193 measured | 29.8%thinking reasoning | 73.4% |
Compare them with other models
Quick answers
Is Qwen3-32B better than Kimi K2 (0905)?
Pick Kimi K2 (0905) if quality matters most: it scores higher (Intelligence Index 15.3 against 8.6). Pick Qwen3-32B if cost matters more: it costs 5.2 times less per token.
Which is cheaper, Qwen3-32B or Kimi K2 (0905)?
Qwen3-32B is cheaper: $0.21 against $1.07 per million tokens, 5.2 times less.
Which is better for reasoning and knowledge, Qwen3-32B or Kimi K2 (0905)?
Kimi K2 (0905) ranks 128 of 198 models on sale for hard questions across science, maths and general knowledge, against 148 for Qwen3-32B.
Which is better for writing, Qwen3-32B or Kimi K2 (0905)?
Kimi K2 (0905) ranks 73 of 188 models on sale for what people prefer in blind comparisons, and judged writing quality, against 139 for Qwen3-32B.
Which is better for sticking to the facts, Qwen3-32B or Kimi K2 (0905)?
Qwen3-32B ranks 22 of 152 models on sale for summarising without inventing things, and factual answers, against 135 for Kimi K2 (0905).
Which has the bigger context window, Qwen3-32B or Kimi K2 (0905)?
Kimi K2 (0905): 262,144 tokens, against 131,072 for Qwen3-32B.
Run Qwen3-32B and Kimi K2 (0905) on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.