Model comparison
Kimi K2 (Jul 2025) vs Mixtral 8x22B
Kimi K2 (Jul 2025) is the stronger model overall, scoring 40.9 to 26.7 on the Noometry Index.
Last verified . 25 shared benchmarks.
Mistral AISummary
- They share 25 benchmarks with published results for both. Kimi K2 (Jul 2025) scores higher in 9 categories and Mixtral 8x22B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Kimi K2 (Jul 2025) leads 62.5 to 37.1.
- The biggest single-benchmark swing is Omni-MATH: 65.4% for Kimi K2 (Jul 2025) and 16.3% for Mixtral 8x22B.
- Kimi K2 (Jul 2025) is cheaper at $0.57 / $2.30 per million input/output tokens, against $2 / $6 for Mixtral 8x22B.
- Kimi K2 (Jul 2025) accepts more context: 262K tokens versus 64K.
Side by side
| Kimi K2 (Jul 2025) | Mixtral 8x22B | |
|---|---|---|
| Provider | Moonshot AI | Mistral AI |
| Noometry Index | 40.9 | 26.7 |
| Released | 2025-07-12 | 2024-04-17 |
| Weights | Open | Open |
| Context window | 262K | 64K |
| Max output | 262K | 64K |
| Input $ / M tokens | $0.57 | $2 |
| Output $ / M tokens | $2.30 | $6 |
| Results tracked | 42 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 43.1 (#83), Mixtral 8x22B: 23.8 (#324)
| Benchmark | Kimi K2 (Jul 2025) | Mixtral 8x22B |
|---|---|---|
| WeirdML | 42.8% | 3.2% |
| LMArena Coding | 1399 | 1166 |
| SWE-bench Verified (bash only) | 63.4% | — |
| Aider Polyglot | 59.1% | — |
| GSO | 4.9% | — |
| BigCodeBench Instruct | — | 40.6% |
| BigCodeBench Complete | — | 50.2% |
| ALE-Bench | 597.5 | — |
| HumanEval+ | — | 72% |
| MBPP+ | — | 64.3% |
Agentic & Tool Use Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 32.2 (#59), Mixtral 8x22B: 22.8 (#119)
| Benchmark | Kimi K2 (Jul 2025) | Mixtral 8x22B |
|---|---|---|
| Terminal-Bench | 35.7% | — |
| Berkeley Function Calling Leaderboard | 59.1% | — |
| Cybench | — | 7.5% |
| METR Time Horizons | 59.2% | — |
Reasoning Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 22.7 (#173), Mixtral 8x22B: 19.1 (#246)
| Benchmark | Kimi K2 (Jul 2025) | Mixtral 8x22B |
|---|---|---|
| LMArena Hard Prompts | 1384 | 1150 |
| Epoch Capabilities Index | 146.02 | 122 |
| ForecastBench | 60.2 | 56.3 |
| SimpleBench | 26.3% | — |
| Kagi LLM Benchmark | 64.4% | — |
| DTBench | — | 55.1% |
Math Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 40.1 (#81), Mixtral 8x22B: 21.2 (#271)
| Benchmark | Kimi K2 (Jul 2025) | Mixtral 8x22B |
|---|---|---|
| Omni-MATH | 65.4% | 16.3% |
| LMArena Math | 1396 | 1184 |
| MATH Level 5 | — | 24.2% |
| FrontierMath (Feb 2025 set) | 21.4% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 36.9 (#153), Mixtral 8x22B: 15.1 (#288)
| Benchmark | Kimi K2 (Jul 2025) | Mixtral 8x22B |
|---|---|---|
| MMLU-Pro | 81.9% | 46% |
| GPQA (HELM) | 65.3% | 33.4% |
| LMArena Expert | 1363 | 1113 |
| GPQA Diamond | — | 34.1% |
| Confabulations | 20.4% | — |
| Vectara Hallucination Rate | 17.9% | — |
| MMLU | — | 77.8% |
Multilingual Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 49.8 (#125), Mixtral 8x22B: 32.9 (#250)
| Benchmark | Kimi K2 (Jul 2025) | Mixtral 8x22B |
|---|---|---|
| LMArena Non-English | 1372 | 1128 |
| LMArena Chinese | 1416 | 1116 |
| LMArena French | 1379 | 1165 |
| LMArena German | 1386 | 1141 |
| LMArena Japanese | 1349 | 1037 |
| LMArena Korean | 1324 | 1056 |
| LMArena Russian | 1384 | 1158 |
| LMArena Spanish | 1383 | 1150 |
Instruction Following Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 71.3 (#151), Mixtral 8x22B: 57.9 (#261)
| Benchmark | Kimi K2 (Jul 2025) | Mixtral 8x22B |
|---|---|---|
| IFEval | 85% | 72.4% |
| LMArena Instruction Following | 1347 | 1147 |
Long Context Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 41.2 (#140), Mixtral 8x22B: 34.6 (#242)
| Benchmark | Kimi K2 (Jul 2025) | Mixtral 8x22B |
|---|---|---|
| LMArena Longer Query | 1353 | 1144 |
| Fiction.LiveBench | 66.7% | — |
| CL-bench | 17.6% | — |
Writing & Preference Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 62.5 (#76), Mixtral 8x22B: 37.1 (#257)
| Benchmark | Kimi K2 (Jul 2025) | Mixtral 8x22B |
|---|---|---|
| LMArena Text | 1380 | 1162 |
| LMArena Creative Writing | 1349 | 1141 |
| WildBench | 86.2% | 71.1% |
| LMArena Multi-Turn | 1371 | 1130 |
| Short-Story Creative Writing | 85.6% | — |
| EQ-Bench Creative Writing | 1666 | — |
Frequently asked questions
Is Kimi K2 (Jul 2025) better than Mixtral 8x22B?
Kimi K2 (Jul 2025) is the stronger model overall, scoring 40.9 to 26.7 on the Noometry Index.
Which is cheaper, Kimi K2 (Jul 2025) or Mixtral 8x22B?
Kimi K2 (Jul 2025) is cheaper. It lists at $0.57 per million input tokens and $2.30 per million output tokens; Mixtral 8x22B lists at $2 and $6.
Is Kimi K2 (Jul 2025) or Mixtral 8x22B better for coding?
Kimi K2 (Jul 2025) scores higher on coding benchmarks: 43.1 versus 23.8 in the Noometry coding category.
Which has the bigger context window?
Kimi K2 (Jul 2025) does, with 262K tokens against 64K.
How many benchmarks do Kimi K2 (Jul 2025) and Mixtral 8x22B share?
25 benchmarks have published results for both models. Kimi K2 (Jul 2025) has 42 scored results on Noometry and Mixtral 8x22B has 34.