Model comparison
Grok-3 mini vs Kimi K2 (Jul 2025)
Grok-3 mini and Kimi K2 (Jul 2025) score almost the same on the Noometry Index (41.2 vs 41.2), so choose on price, context window or the category you care about most.
Last verified . 30 shared benchmarks.
Summary
- They share 30 benchmarks with published results for both. Grok-3 mini scores higher in 2 categories and Kimi K2 (Jul 2025) in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Kimi K2 (Jul 2025) leads 62.3 to 52.5.
- The biggest single-benchmark swing is Omni-MATH: 31.8% for Grok-3 mini and 65.4% for Kimi K2 (Jul 2025).
- Kimi K2 (Jul 2025) has downloadable open weights; the other is API-only.
Side by side
| Grok-3 mini | Kimi K2 (Jul 2025) | |
|---|---|---|
| Provider | xAI | Moonshot AI |
| Noometry Index | 41.2 | 41.2 |
| Released | 2025-04-09 | 2025-07-12 |
| Weights | Proprietary | Open |
| Context window | — | 262K |
| Max output | — | 262K |
| Input $ / M tokens | — | $0.57 |
| Output $ / M tokens | — | $2.30 |
| Results tracked | 35 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2 (Jul 2025) leads
Grok-3 mini: 40.8 (#131), Kimi K2 (Jul 2025): 42.4 (#102)
| Benchmark | Grok-3 mini | Kimi K2 (Jul 2025) |
|---|---|---|
| Aider Polyglot | 49.3% | 59.1% |
| WeirdML | 42.6% | 42.8% |
| LMArena Coding | 1379 | 1399 |
| SWE-bench Verified (bash only) | — | 63.4% |
| GSO | — | 4.9% |
| ALE-Bench | — | 597.5 |
Agentic & Tool Use Not comparable
Grok-3 mini: —, Kimi K2 (Jul 2025): 32.4 (#64)
| Benchmark | Grok-3 mini | Kimi K2 (Jul 2025) |
|---|---|---|
| Terminal-Bench | — | 35.7% |
| Berkeley Function Calling Leaderboard | — | 59.1% |
| METR Time Horizons | — | 59.2% |
Reasoning Kimi K2 (Jul 2025) leads
Grok-3 mini: 13.6 (#334), Kimi K2 (Jul 2025): 23.3 (#179)
| Benchmark | Grok-3 mini | Kimi K2 (Jul 2025) |
|---|---|---|
| Kagi LLM Benchmark | 61.3% | 64.4% |
| LMArena Hard Prompts | 1375 | 1384 |
| Epoch Capabilities Index | 140.35 | 146.01 |
| ARC-AGI-2 | 0.4% | — |
| SimpleBench | — | 26.3% |
| ARC-AGI-1 | 16.5% | — |
| ForecastBench | — | 60.2 |
Math Too close to call
Grok-3 mini: 42.1 (#85), Kimi K2 (Jul 2025): 42.7 (#83)
| Benchmark | Grok-3 mini | Kimi K2 (Jul 2025) |
|---|---|---|
| Omni-MATH | 31.8% | 65.4% |
| LMArena Math | 1386 | 1397 |
| FrontierMath (Feb 2025 set) | 5.9% | 21.4% |
| OTIS Mock AIME 2024-2025 | 77.8% | — |
| MATH Level 5 | 90.9% | — |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Grok-3 mini leads
Grok-3 mini: 46.4 (#81), Kimi K2 (Jul 2025): 37.3 (#157)
| Benchmark | Grok-3 mini | Kimi K2 (Jul 2025) |
|---|---|---|
| MMLU-Pro | 79.9% | 81.9% |
| Confabulations | 10.8% | 20.4% |
| GPQA (HELM) | 67.5% | 65.3% |
| LMArena Expert | 1395 | 1365 |
| GPQA Diamond | 76.3% | — |
| Vectara Hallucination Rate | — | 17.9% |
Multilingual Kimi K2 (Jul 2025) leads
Grok-3 mini: 48.1 (#145), Kimi K2 (Jul 2025): 49.6 (#130)
| Benchmark | Grok-3 mini | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1352 | 1372 |
| LMArena Chinese | 1387 | 1415 |
| LMArena French | 1357 | 1379 |
| LMArena German | 1349 | 1387 |
| LMArena Japanese | 1342 | 1349 |
| LMArena Korean | 1335 | 1325 |
| LMArena Russian | 1353 | 1385 |
| LMArena Spanish | 1381 | 1386 |
Instruction Following Grok-3 mini leads
Grok-3 mini: 78.5 (#9), Kimi K2 (Jul 2025): 71.1 (#156)
| Benchmark | Grok-3 mini | Kimi K2 (Jul 2025) |
|---|---|---|
| IFEval | 95.1% | 85% |
| LMArena Instruction Following | 1357 | 1348 |
Long Context Too close to call
Grok-3 mini: 41.0 (#147), Kimi K2 (Jul 2025): 41.2 (#145)
| Benchmark | Grok-3 mini | Kimi K2 (Jul 2025) |
|---|---|---|
| Fiction.LiveBench | 66.7% | 66.7% |
| LMArena Longer Query | 1372 | 1353 |
| CL-bench | — | 17.6% |
Writing & Preference Kimi K2 (Jul 2025) leads
Grok-3 mini: 52.5 (#169), Kimi K2 (Jul 2025): 62.3 (#78)
| Benchmark | Grok-3 mini | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1370 | 1380 |
| LMArena Creative Writing | 1342 | 1350 |
| Short-Story Creative Writing | 73.5% | 85.6% |
| WildBench | 65.1% | 86.2% |
| LMArena Multi-Turn | 1355 | 1371 |
| EQ-Bench Creative Writing | — | 1666 |
Frequently asked questions
Is Grok-3 mini better than Kimi K2 (Jul 2025)?
Grok-3 mini and Kimi K2 (Jul 2025) score almost the same on the Noometry Index (41.2 vs 41.2), so choose on price, context window or the category you care about most.
Is Grok-3 mini or Kimi K2 (Jul 2025) better for coding?
Kimi K2 (Jul 2025) scores higher on coding benchmarks: 42.4 versus 40.8 in the Noometry coding category.
How many benchmarks do Grok-3 mini and Kimi K2 (Jul 2025) share?
30 benchmarks have published results for both models. Grok-3 mini has 35 scored results on Noometry and Kimi K2 (Jul 2025) has 42.