Model comparison
Grok 4.6 vs Kimi K2 (Jul 2025)
Grok 4.6 is the stronger model overall, scoring 56.9 to 41.2 on the Noometry Index. Kimi K2 (Jul 2025) costs 3.0× less per token, which makes it the better buy when Grok 4.6's lead doesn't matter for your workload.
Last verified . 21 shared benchmarks.
Summary
- They share 21 benchmarks with published results for both. Grok 4.6 scores higher in 8 categories and Kimi K2 (Jul 2025) in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.6 leads 61.4 to 23.3.
- The biggest single-benchmark swing is SimpleBench: 75.9% for Grok 4.6 and 26.3% for Kimi K2 (Jul 2025).
- Kimi K2 (Jul 2025) is cheaper at $0.57 / $2.30 per million input/output tokens, against $2 / $6 for Grok 4.6.
- Grok 4.6 accepts more context: 500K tokens versus 262K.
- Kimi K2 (Jul 2025) has downloadable open weights; the other is API-only.
Side by side
| Grok 4.6 | Kimi K2 (Jul 2025) | |
|---|---|---|
| Provider | xAI | Moonshot AI |
| Noometry Index | 56.9 | 41.2 |
| Released | 2026-08-12 | 2025-07-12 |
| Weights | Proprietary | Open |
| Context window | 500K | 262K |
| Max output | 500K | 262K |
| Input $ / M tokens | $2 | $0.57 |
| Output $ / M tokens | $6 | $2.30 |
| Results tracked | 49 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.6 leads
Grok 4.6: 58.5 (#16), Kimi K2 (Jul 2025): 42.4 (#102)
| Benchmark | Grok 4.6 | Kimi K2 (Jul 2025) |
|---|---|---|
| WeirdML | 67.3% | 42.8% |
| LMArena Coding | 1465 | 1399 |
| ALE-Bench | 1,508 | 597.5 |
| DeepSWE | 67.5% | — |
| FrontierCode | 48% | — |
| SWE-bench Verified (bash only) | — | 63.4% |
| Aider Polyglot | — | 59.1% |
| CursorBench | 41.4% | — |
| LMArena WebDev | 1617 | — |
| FrontierSWE | 25.3% | — |
| SciCode | 56.5% | — |
| GSO | — | 4.9% |
Agentic & Tool Use Grok 4.6 leads
Grok 4.6: 39.4 (#27), Kimi K2 (Jul 2025): 32.4 (#64)
| Benchmark | Grok 4.6 | Kimi K2 (Jul 2025) |
|---|---|---|
| Terminal-Bench | — | 35.7% |
| APEX-Agents | 65.3% | — |
| Berkeley Function Calling Leaderboard | — | 59.1% |
| GDP.pdf | 17.2% | — |
| METR Time Horizons | — | 59.2% |
| Vending-Bench 2 | 9,047 | — |
Reasoning Grok 4.6 leads
Grok 4.6: 61.4 (#20), Kimi K2 (Jul 2025): 23.3 (#179)
| Benchmark | Grok 4.6 | Kimi K2 (Jul 2025) |
|---|---|---|
| SimpleBench | 75.9% | 26.3% |
| LMArena Hard Prompts | 1447 | 1384 |
| Epoch Capabilities Index | 156.44 | 146.01 |
| ARC-AGI-2 | 67.1% | — |
| Kagi LLM Benchmark | — | 64.4% |
| NYT Connections (extended) | 80% | — |
| ARC-AGI-1 | 87.5% | — |
| CritPt | 19.7% | — |
| Chess Puzzles | 40% | — |
| EBR-Bench | 30.5% | — |
| Mystery Game Puzzles | 34% | — |
| DTBench | 97.3% | — |
| LMCA | 48.5% | — |
| ForecastBench | — | 60.2 |
Math Grok 4.6 leads
Grok 4.6: 67.0 (#24), Kimi K2 (Jul 2025): 42.7 (#83)
| Benchmark | Grok 4.6 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Math | 1423 | 1397 |
| FrontierMath (Tiers 1-3) | 66% | — |
| FrontierMath Tier 4 | 31.7% | — |
| OTIS Mock AIME 2024-2025 | 99.2% | — |
| ProofBench | 51% | — |
| Omni-MATH | — | 65.4% |
| FrontierMath (Feb 2025 set) | — | 21.4% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Grok 4.6 leads
Grok 4.6: 63.3 (#20), Kimi K2 (Jul 2025): 37.3 (#157)
| Benchmark | Grok 4.6 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Expert | 1467 | 1365 |
| GPQA Diamond | 94% | — |
| SimpleQA Verified | 49.3% | — |
| MMLU-Pro | — | 81.9% |
| Confabulations | — | 20.4% |
| Vectara Hallucination Rate | — | 17.9% |
| GPQA (HELM) | — | 65.3% |
Multimodal Not comparable
Grok 4.6: 43.6 (#23), Kimi K2 (Jul 2025): —
| Benchmark | Grok 4.6 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Vision | 1263 | — |
| Blueprint-Bench 2 | 33.2% | — |
| Furniture Assembly | 40% | — |
| LMArena Document | 1452 | — |
Multilingual Grok 4.6 leads
Grok 4.6: 53.0 (#74), Kimi K2 (Jul 2025): 49.6 (#130)
| Benchmark | Grok 4.6 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1420 | 1372 |
| LMArena Chinese | 1480 | 1415 |
| LMArena French | 1461 | 1379 |
| LMArena German | 1431 | 1387 |
| LMArena Japanese | 1376 | 1349 |
| LMArena Korean | 1397 | 1325 |
| LMArena Russian | 1422 | 1385 |
| LMArena Spanish | 1404 | 1386 |
Instruction Following Grok 4.6 leads
Grok 4.6: 75.4 (#63), Kimi K2 (Jul 2025): 71.1 (#156)
| Benchmark | Grok 4.6 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Instruction Following | 1431 | 1348 |
| IFEval | — | 85% |
Long Context Grok 4.6 leads
Grok 4.6: 44.5 (#66), Kimi K2 (Jul 2025): 41.2 (#145)
| Benchmark | Grok 4.6 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Longer Query | 1454 | 1353 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 17.6% |
Writing & Preference Too close to call
Grok 4.6: 62.3 (#80), Kimi K2 (Jul 2025): 62.3 (#78)
| Benchmark | Grok 4.6 | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1428 | 1380 |
| LMArena Creative Writing | 1428 | 1350 |
| LMArena Multi-Turn | 1425 | 1371 |
| Short-Story Creative Writing | — | 85.6% |
| EQ-Bench Creative Writing | — | 1666 |
| WildBench | — | 86.2% |
Frequently asked questions
Is Grok 4.6 better than Kimi K2 (Jul 2025)?
Grok 4.6 is the stronger model overall, scoring 56.9 to 41.2 on the Noometry Index. Kimi K2 (Jul 2025) costs 3.0× less per token, which makes it the better buy when Grok 4.6's lead doesn't matter for your workload.
Which is cheaper, Grok 4.6 or Kimi K2 (Jul 2025)?
Kimi K2 (Jul 2025) is cheaper. It lists at $0.57 per million input tokens and $2.30 per million output tokens; Grok 4.6 lists at $2 and $6.
Is Grok 4.6 or Kimi K2 (Jul 2025) better for coding?
Grok 4.6 scores higher on coding benchmarks: 58.5 versus 42.4 in the Noometry coding category.
Which has the bigger context window?
Grok 4.6 does, with 500K tokens against 262K.
How many benchmarks do Grok 4.6 and Kimi K2 (Jul 2025) share?
21 benchmarks have published results for both models. Grok 4.6 has 49 scored results on Noometry and Kimi K2 (Jul 2025) has 42.