Up to four models
Kimi K2.5 vs Qwen 3.5 Flash (hosted 35B-A3B)
On the index (GPT-4 = 100), Kimi K2.5 scores 145 and Qwen 3.5 Flash (hosted 35B-A3B) 138. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
Kimi K2.5
145130–161
WeightsOpen
ReleasedJan 2026
Price in / out$0.50 / $2.5
Context262K
Core tests taken5 of 9
138121–155
WeightsClosed
ReleasedFeb 2026
Price in / out–
Context–
Core tests taken2 of 9
Kimi K2.5 and Qwen 3.5 Flash (hosted 35B-A3B) are too close to call (7 points apart, within the uncertainty), ahead on 1 of 1 shared core tests.
Test by testthird-party results
Kimi K2.5
Qwen 3.5 Flash (hosted 35B-A3B)
Every result as a table
| Test | Kimi K2.5 | Qwen 3.5 Flash (hosted 35B-A3B) |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 87.6% | 82.3% |
| GPQA Diamond (Vals AI) | 84.1% | 82.8% |
| MMLU Pro | 85.9% | 84.1% |
| MMMU Pro | 84.3% | 81.9% |
| SimpleQA Verified | 34.3% | 20.3% |
| Maths | ||
| AIME-style maths (OTIS mock) | 92.2% | 84.4% |
| FrontierMath (2025 set) | 27.9% | 6.2% |
| FrontierMath tier 4 (2025 set) | 4.2% | 0.0% |
| Coding | ||
| LiveCodeBench | 83.9% | 83.3% |
| SWE-bench | 70.0% | 64.4% |
| Professional work | ||
| CorpFin v2 | 68.3% | 63.6% |
| MedCode | 39.3% | 33.0% |
| MedScribe | 76.4% | 70.6% |
| MortgageTax | 66.5% | 67.4% |
| TaxEval v2 | 74.2% | 72.2% |
Tests at least two of these models have taken. Bold rows are the core tests.