Model comparison
GPT-5 Pro vs Kimi K2.6
Kimi K2.6 is the stronger model overall, scoring 47.7 to 46.4 on the Noometry Index.
Last verified . 5 shared benchmarks.
Summary
- They share 5 benchmarks with published results for both. GPT-5 Pro scores higher in 1 category and Kimi K2.6 in 3 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.6 leads 57.0 to 48.5.
- The biggest single-benchmark swing is FrontierMath Tier 4: 19.5% for GPT-5 Pro and 25.6% for Kimi K2.6.
- Kimi K2.6 is cheaper at $0.95 / $4 per million input/output tokens, against $15 / $120 for GPT-5 Pro.
- GPT-5 Pro accepts more context: 400K tokens versus 262K.
- Kimi K2.6 has downloadable open weights; the other is API-only.
Side by side
| GPT-5 Pro | Kimi K2.6 | |
|---|---|---|
| Provider | OpenAI | Moonshot AI |
| Noometry Index | 46.4 | 47.7 |
| Released | 2025-10-06 | 2026-04-20 |
| Weights | Proprietary | Open |
| Context window | 400K | 262K |
| Max output | 272K | 262K |
| Input $ / M tokens | $15 | $0.95 |
| Output $ / M tokens | $120 | $4 |
| Results tracked | 12 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.6 leads
GPT-5 Pro: 44.0 (#80), Kimi K2.6: 50.7 (#43)
| Benchmark | GPT-5 Pro | Kimi K2.6 |
|---|---|---|
| WeirdML | 60.4% | 55.9% |
| SWE-bench Verified | — | 76.7% |
| LMArena WebDev | — | 1509 |
| SciCode | — | 53.5% |
| LMArena Coding | — | 1488 |
| ALE-Bench | — | 1,093 |
| AlgoTune | 1.31 | — |
Agentic & Tool Use Not comparable
GPT-5 Pro: —, Kimi K2.6: 21.9 (#137)
| Benchmark | GPT-5 Pro | Kimi K2.6 |
|---|---|---|
| OSWorld 2.0 | — | 4.6% |
| ExploitBench | — | 18.4% |
| GBAEval | — | 0.9% |
| GDP.pdf | — | 12% |
| Vending-Bench 2 | — | 6,205 |
Reasoning Kimi K2.6 leads
GPT-5 Pro: 38.9 (#62), Kimi K2.6: 40.5 (#55)
| Benchmark | GPT-5 Pro | Kimi K2.6 |
|---|---|---|
| Epoch Capabilities Index | 150.28 | 151.05 |
| ARC-AGI-2 | 18.3% | — |
| SimpleBench | 61.6% | — |
| Kagi LLM Benchmark | 76.8% | — |
| NYT Connections (extended) | — | 87.2% |
| ARC-AGI-1 | 70.2% | — |
| CritPt | — | 8% |
| Chess Puzzles | — | 26% |
| EnigmaEval | 18.8% | — |
| EBR-Bench | — | 2.4% |
| LMArena Hard Prompts | — | 1470 |
| Mystery Game Puzzles | — | 18% |
| DTBench | — | 90.9% |
| LMCA | — | 37.3% |
Math Kimi K2.6 leads
GPT-5 Pro: 48.5 (#63), Kimi K2.6: 57.0 (#41)
| Benchmark | GPT-5 Pro | Kimi K2.6 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 55.8% | 57.2% |
| FrontierMath Tier 4 | 19.5% | 25.6% |
| FrontierMath Tier 4 (v1) | 14.6% | 14.6% |
| MathArena Final-Answer Competitions | — | 72.9% |
| OTIS Mock AIME 2024-2025 | — | 96.1% |
| ProofBench | — | 16% |
| LMArena Math | — | 1475 |
| FrontierMath (Feb 2025 set) | — | 39% |
Knowledge GPT-5 Pro leads
GPT-5 Pro: 56.7 (#42), Kimi K2.6: 54.0 (#54)
| Benchmark | GPT-5 Pro | Kimi K2.6 |
|---|---|---|
| GPQA Diamond | — | 90.8% |
| Humanity's Last Exam | 31.6% | — |
| SimpleQA Verified | — | 34.9% |
| Vectara Hallucination Rate | — | 10.8% |
| LMArena Expert | — | 1491 |
Multimodal Not comparable
GPT-5 Pro: —, Kimi K2.6: 31.6 (#103)
| Benchmark | GPT-5 Pro | Kimi K2.6 |
|---|---|---|
| LMArena Vision | — | 1283 |
| Blueprint-Bench 2 | — | 3.9% |
| Furniture Assembly | — | 21.7% |
| LMArena Document | — | 1451 |
Multilingual Not comparable
GPT-5 Pro: —, Kimi K2.6: 54.9 (#37)
| Benchmark | GPT-5 Pro | Kimi K2.6 |
|---|---|---|
| LMArena Non-English | — | 1446 |
| LMArena Chinese | — | 1521 |
| LMArena French | — | 1471 |
| LMArena German | — | 1450 |
| LMArena Japanese | — | 1443 |
| LMArena Korean | — | 1427 |
| LMArena Russian | — | 1446 |
| LMArena Spanish | — | 1464 |
Instruction Following Not comparable
GPT-5 Pro: —, Kimi K2.6: 76.3 (#43)
| Benchmark | GPT-5 Pro | Kimi K2.6 |
|---|---|---|
| LMArena Instruction Following | — | 1451 |
Long Context Not comparable
GPT-5 Pro: —, Kimi K2.6: 44.9 (#52)
| Benchmark | GPT-5 Pro | Kimi K2.6 |
|---|---|---|
| LMArena Longer Query | — | 1468 |
Writing & Preference Not comparable
GPT-5 Pro: —, Kimi K2.6: 68.5 (#26)
| Benchmark | GPT-5 Pro | Kimi K2.6 |
|---|---|---|
| LMArena Text | — | 1455 |
| LMArena Creative Writing | — | 1434 |
| EQ-Bench Creative Writing | — | 1725 |
| EQ-Bench 4 | — | 1202 |
| LMArena Multi-Turn | — | 1453 |
Frequently asked questions
Is GPT-5 Pro better than Kimi K2.6?
Kimi K2.6 is the stronger model overall, scoring 47.7 to 46.4 on the Noometry Index.
Which is cheaper, GPT-5 Pro or Kimi K2.6?
Kimi K2.6 is cheaper. It lists at $0.95 per million input tokens and $4 per million output tokens; GPT-5 Pro lists at $15 and $120.
Is GPT-5 Pro or Kimi K2.6 better for coding?
Kimi K2.6 scores higher on coding benchmarks: 50.7 versus 44.0 in the Noometry coding category.
Which has the bigger context window?
GPT-5 Pro does, with 400K tokens against 262K.
How many benchmarks do GPT-5 Pro and Kimi K2.6 share?
5 benchmarks have published results for both models. GPT-5 Pro has 12 scored results on Noometry and Kimi K2.6 has 51.