Model comparison
Kimi K2.6 vs o3
Kimi K2.6 and o3 score almost the same on the Noometry Index (47.7 vs 47.5), so choose on price, context window or the category you care about most.
Last verified . 34 shared benchmarks.
Summary
- They share 34 benchmarks with published results for both. Kimi K2.6 scores higher in 6 categories and o3 in 4 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where o3 leads 34.5 to 21.9.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 57.2% for Kimi K2.6 and 33.3% for o3.
- Kimi K2.6 is cheaper at $0.95 / $4 per million input/output tokens, against $2 / $8 for o3.
- Kimi K2.6 accepts more context: 262K tokens versus 200K.
- Kimi K2.6 has downloadable open weights; the other is API-only.
Side by side
| Kimi K2.6 | o3 | |
|---|---|---|
| Provider | Moonshot AI | OpenAI |
| Noometry Index | 47.7 | 47.5 |
| Released | 2026-04-20 | 2025-04-16 |
| Weights | Open | Proprietary |
| Context window | 262K | 200K |
| Max output | 262K | 100K |
| Input $ / M tokens | $0.95 | $2 |
| Output $ / M tokens | $4 | $8 |
| Results tracked | 51 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.6 leads
Kimi K2.6: 50.7 (#43), o3: 46.8 (#64)
| Benchmark | Kimi K2.6 | o3 |
|---|---|---|
| SWE-bench Verified | 76.7% | 62.3% |
| WeirdML | 55.9% | 52.4% |
| LMArena Coding | 1488 | 1408 |
| ALE-Bench | 1,093 | 933.55 |
| SWE-bench Verified (bash only) | — | 58.4% |
| Aider Polyglot | — | 81.3% |
| LMArena WebDev | 1509 | — |
| SciCode | 53.5% | — |
| GSO | — | 8.8% |
| CadEval | — | 74% |
Agentic & Tool Use o3 leads
Kimi K2.6: 21.9 (#137), o3: 34.5 (#44)
| Benchmark | Kimi K2.6 | o3 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 63% |
| OSWorld 2.0 | 4.6% | — |
| GDPval | — | 30.8% |
| DeepResearch Bench | — | 45.2% |
| OSWorld | — | 23% |
| ExploitBench | 18.4% | — |
| GBAEval | 0.9% | — |
| GDP.pdf | 12% | — |
| LMArena Search | — | 1144 |
| METR Time Horizons | — | 65.4% |
| Vending-Bench 2 | 6,205 | — |
Reasoning Kimi K2.6 leads
Kimi K2.6: 40.5 (#55), o3: 32.0 (#78)
| Benchmark | Kimi K2.6 | o3 |
|---|---|---|
| CritPt | 8% | 1.4% |
| Chess Puzzles | 26% | 38% |
| LMArena Hard Prompts | 1470 | 1402 |
| Mystery Game Puzzles | 18% | 29% |
| DTBench | 90.9% | 84.8% |
| LMCA | 37.3% | 39.7% |
| Epoch Capabilities Index | 151.05 | 146.86 |
| ARC-AGI-2 | — | 6.5% |
| SimpleBench | — | 53.1% |
| Kagi LLM Benchmark | — | 67.6% |
| NYT Connections (extended) | 87.2% | — |
| ARC-AGI-1 | — | 60.8% |
| EnigmaEval | — | 13.1% |
| EBR-Bench | 2.4% | — |
| ForecastBench | — | 62.5 |
Math Kimi K2.6 leads
Kimi K2.6: 57.0 (#41), o3: 50.2 (#58)
| Benchmark | Kimi K2.6 | o3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 57.2% | 33.3% |
| OTIS Mock AIME 2024-2025 | 96.1% | 84.4% |
| LMArena Math | 1475 | 1426 |
| FrontierMath (Feb 2025 set) | 39% | 18.7% |
| FrontierMath Tier 4 (v1) | 14.6% | 2.1% |
| FrontierMath Tier 4 | 25.6% | — |
| MathArena Final-Answer Competitions | 72.9% | — |
| ProofBench | 16% | — |
| Omni-MATH | — | 71.4% |
| MATH Level 5 | — | 97.8% |
Knowledge Too close to call
Kimi K2.6: 54.0 (#54), o3: 54.6 (#52)
| Benchmark | Kimi K2.6 | o3 |
|---|---|---|
| GPQA Diamond | 90.8% | 81.8% |
| SimpleQA Verified | 34.9% | 49.4% |
| LMArena Expert | 1491 | 1402 |
| Humanity's Last Exam | — | 20.3% |
| MMLU-Pro | — | 85.9% |
| Confabulations | — | 14.4% |
| Vectara Hallucination Rate | 10.8% | — |
| GPQA (HELM) | — | 75.3% |
Multimodal o3 leads
Kimi K2.6: 31.6 (#103), o3: 41.4 (#36)
| Benchmark | Kimi K2.6 | o3 |
|---|---|---|
| LMArena Vision | 1283 | 1214 |
| GeoBench | — | 74% |
| VPCT | — | 52% |
| Blueprint-Bench 2 | 3.9% | — |
| Furniture Assembly | 21.7% | — |
| LMArena Document | 1451 | — |
Multilingual Kimi K2.6 leads
Kimi K2.6: 54.9 (#37), o3: 51.7 (#105)
| Benchmark | Kimi K2.6 | o3 |
|---|---|---|
| LMArena Non-English | 1446 | 1401 |
| LMArena Chinese | 1521 | 1437 |
| LMArena French | 1471 | 1430 |
| LMArena German | 1450 | 1420 |
| LMArena Japanese | 1443 | 1403 |
| LMArena Korean | 1427 | 1370 |
| LMArena Russian | 1446 | 1406 |
| LMArena Spanish | 1464 | 1395 |
Instruction Following Kimi K2.6 leads
Kimi K2.6: 76.3 (#43), o3: 72.8 (#127)
| Benchmark | Kimi K2.6 | o3 |
|---|---|---|
| LMArena Instruction Following | 1451 | 1368 |
| IFEval | — | 86.9% |
Long Context o3 leads
Kimi K2.6: 44.9 (#52), o3: 53.3 (#6)
| Benchmark | Kimi K2.6 | o3 |
|---|---|---|
| LMArena Longer Query | 1468 | 1372 |
| Fiction.LiveBench | — | 88.9% |
| CL-bench | — | 17.8% |
Writing & Preference Kimi K2.6 leads
Kimi K2.6: 68.5 (#26), o3: 63.5 (#64)
| Benchmark | Kimi K2.6 | o3 |
|---|---|---|
| LMArena Text | 1455 | 1410 |
| LMArena Creative Writing | 1434 | 1359 |
| EQ-Bench Creative Writing | 1725 | 1676 |
| LMArena Multi-Turn | 1453 | 1405 |
| Short-Story Creative Writing | — | 83.9% |
| WildBench | — | 86.1% |
| EQ-Bench 4 | 1202 | — |
Frequently asked questions
Is Kimi K2.6 better than o3?
Kimi K2.6 and o3 score almost the same on the Noometry Index (47.7 vs 47.5), so choose on price, context window or the category you care about most.
Which is cheaper, Kimi K2.6 or o3?
Kimi K2.6 is cheaper. It lists at $0.95 per million input tokens and $4 per million output tokens; o3 lists at $2 and $8.
Is Kimi K2.6 or o3 better for coding?
Kimi K2.6 scores higher on coding benchmarks: 50.7 versus 46.8 in the Noometry coding category.
Which has the bigger context window?
Kimi K2.6 does, with 262K tokens against 200K.
How many benchmarks do Kimi K2.6 and o3 share?
34 benchmarks have published results for both models. Kimi K2.6 has 51 scored results on Noometry and o3 has 63.