Model comparison
Grok 3 vs Kimi K2 Thinking Turbo
Kimi K2 Thinking Turbo is the stronger model overall, scoring 45.8 to 39.9 on the Noometry Index.
Last verified . 19 shared benchmarks.
Summary
- They share 19 benchmarks with published results for both. Grok 3 scores higher in 3 categories and Kimi K2 Thinking Turbo in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Kimi K2 Thinking Turbo leads 32.5 to 13.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 55.6% for Grok 3 and 83.1% for Kimi K2 Thinking Turbo.
- Kimi K2 Thinking Turbo has downloadable open weights; the other is API-only.
Side by side
| Grok 3 | Kimi K2 Thinking Turbo | |
|---|---|---|
| Provider | xAI | Moonshot AI |
| Noometry Index | 39.9 | 45.8 |
| Released | 2025-04-09 | 2025-11-06 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 40 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 3 leads
Grok 3: 41.9 (#115), Kimi K2 Thinking Turbo: 38.4 (#178)
| Benchmark | Grok 3 | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Coding | 1432 | 1454 |
| Aider Polyglot | 53.3% | — |
| LMArena WebDev | — | 1322 |
| WeirdML | 37.2% | — |
Agentic & Tool Use Not comparable
Grok 3: 30.5 (#76), Kimi K2 Thinking Turbo: —
| Benchmark | Grok 3 | Kimi K2 Thinking Turbo |
|---|---|---|
| BALROG | 29.5% | — |
Reasoning Kimi K2 Thinking Turbo leads
Grok 3: 13.7 (#333), Kimi K2 Thinking Turbo: 32.5 (#75)
| Benchmark | Grok 3 | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Hard Prompts | 1434 | 1428 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 36.1% | — |
| Kagi LLM Benchmark | 61.3% | — |
| ARC-AGI-1 | 5.5% | — |
| Chess Puzzles | — | 20% |
| Epoch Capabilities Index | 138.33 | — |
Math Kimi K2 Thinking Turbo leads
Grok 3: 38.0 (#145), Kimi K2 Thinking Turbo: 47.4 (#68)
| Benchmark | Grok 3 | Kimi K2 Thinking Turbo |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 55.6% | 83.1% |
| LMArena Math | 1391 | 1429 |
| Omni-MATH | 46.4% | — |
| MATH Level 5 | 88.7% | — |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Kimi K2 Thinking Turbo leads
Grok 3: 46.2 (#82), Kimi K2 Thinking Turbo: 50.9 (#69)
| Benchmark | Grok 3 | Kimi K2 Thinking Turbo |
|---|---|---|
| GPQA Diamond | 75.8% | 84.2% |
| LMArena Expert | 1421 | 1439 |
| MMLU-Pro | 78.8% | — |
| Confabulations | 14.2% | — |
| Vectara Hallucination Rate | 5.8% | — |
| GPQA (HELM) | 65% | — |
Multilingual Too close to call
Grok 3: 52.3 (#87), Kimi K2 Thinking Turbo: 51.4 (#109)
| Benchmark | Grok 3 | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Non-English | 1410 | 1398 |
| LMArena Chinese | 1448 | 1456 |
| LMArena French | 1460 | 1425 |
| LMArena German | 1431 | 1390 |
| LMArena Japanese | 1387 | 1357 |
| LMArena Korean | 1373 | 1330 |
| LMArena Russian | 1416 | 1391 |
| LMArena Spanish | 1417 | 1406 |
Instruction Following Grok 3 leads
Grok 3: 75.0 (#73), Kimi K2 Thinking Turbo: 74.0 (#109)
| Benchmark | Grok 3 | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Instruction Following | 1409 | 1403 |
| IFEval | 88.4% | — |
Long Context Kimi K2 Thinking Turbo leads
Grok 3: 38.7 (#192), Kimi K2 Thinking Turbo: 43.2 (#102)
| Benchmark | Grok 3 | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Longer Query | 1439 | 1415 |
| Fiction.LiveBench | 58.3% | — |
Writing & Preference Kimi K2 Thinking Turbo leads
Grok 3: 55.8 (#141), Kimi K2 Thinking Turbo: 60.0 (#104)
| Benchmark | Grok 3 | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Text | 1426 | 1415 |
| LMArena Creative Writing | 1414 | 1374 |
| LMArena Multi-Turn | 1425 | 1414 |
| Short-Story Creative Writing | 76.4% | — |
| EQ-Bench Creative Writing | 1186 | — |
| WildBench | 84.9% | — |
Frequently asked questions
Is Grok 3 better than Kimi K2 Thinking Turbo?
Kimi K2 Thinking Turbo is the stronger model overall, scoring 45.8 to 39.9 on the Noometry Index.
Is Grok 3 or Kimi K2 Thinking Turbo better for coding?
Grok 3 scores higher on coding benchmarks: 41.9 versus 38.4 in the Noometry coding category.
How many benchmarks do Grok 3 and Kimi K2 Thinking Turbo share?
19 benchmarks have published results for both models. Grok 3 has 40 scored results on Noometry and Kimi K2 Thinking Turbo has 21.