Model comparison
GPT-5.1 vs Kimi K3
Kimi K3 is the stronger model overall, scoring 59.5 to 49.0 on the Noometry Index. GPT-5.1 costs 1.7× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Last verified . 35 shared benchmarks.
Summary
- They share 35 benchmarks with published results for both. GPT-5.1 scores higher in 3 categories and Kimi K3 in 7 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Kimi K3 leads 63.0 to 39.8.
- The biggest single-benchmark swing is ARC-AGI-2: 17.6% for GPT-5.1 and 60.4% for Kimi K3.
- GPT-5.1 is cheaper at $1.25 / $10 per million input/output tokens, against $3 / $15 for Kimi K3.
- Kimi K3 accepts more context: 1.05M tokens versus 400K.
- Kimi K3 has downloadable open weights; the other is API-only.
Side by side
| GPT-5.1 | Kimi K3 | |
|---|---|---|
| Provider | OpenAI | Moonshot AI |
| Noometry Index | 49.0 | 59.5 |
| Released | 2025-11-13 | 2026-07-16 |
| Weights | Proprietary | Open |
| Context window | 400K | 1.05M |
| Max output | 128K | 1.05M |
| Input $ / M tokens | $1.25 | $3 |
| Output $ / M tokens | $10 | $15 |
| Results tracked | 63 | 53 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
GPT-5.1: 46.4 (#66), Kimi K3: 61.0 (#10)
| Benchmark | GPT-5.1 | Kimi K3 |
|---|---|---|
| LMArena WebDev | 1395 | 1654 |
| SciCode | 43.3% | 59.5% |
| WeirdML | 60.8% | 82.6% |
| LMArena Coding | 1454 | 1508 |
| ALE-Bench | 1,192 | 1,524 |
| SWE-bench Verified | 68% | — |
| DeepSWE | — | 68.5% |
| FrontierCode | — | 44.2% |
| SWE-bench Verified (bash only) | 66% | — |
| FrontierSWE | — | 25.9% |
| GSO | 13.7% | — |
| LiveBench Coding | 72.5% | — |
Agentic & Tool Use Kimi K3 leads
GPT-5.1: 32.7 (#60), Kimi K3: 41.8 (#20)
| Benchmark | GPT-5.1 | Kimi K3 |
|---|---|---|
| Vending-Bench 2 | 1,473 | 5,165 |
| Terminal-Bench | 47.6% | — |
| APEX-Agents | — | 50.6% |
| τ²-bench Banking | — | 37.1% |
| DeepResearch Bench | 42.8% | — |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| GDP.pdf | — | 19% |
| LMArena Search | 1199 | — |
Reasoning Kimi K3 leads
GPT-5.1: 39.8 (#58), Kimi K3: 63.0 (#17)
| Benchmark | GPT-5.1 | Kimi K3 |
|---|---|---|
| ARC-AGI-2 | 17.6% | 60.4% |
| SimpleBench | 53.2% | 60.7% |
| ARC-AGI-1 | 72.8% | 94.5% |
| CritPt | 4.9% | 23.4% |
| Chess Puzzles | 32% | 39% |
| LMArena Hard Prompts | 1457 | 1496 |
| Mystery Game Puzzles | 19% | 26% |
| DTBench | 90.1% | 91.2% |
| LMCA | 43.9% | 52.7% |
| Epoch Capabilities Index | 149.64 | 157.45 |
| ForecastBench | 58.1 | 61.1 |
| NYT Connections (extended) | — | 93.6% |
| EnigmaEval | 11.2% | — |
| LiveBench Reasoning | 95.8% | — |
| LiveBench Data Analysis | 72.1% | — |
| Surface Evolver Bench | — | 95% |
| LiveBench | 78.8% | — |
Math Kimi K3 leads
GPT-5.1: 52.2 (#51), Kimi K3: 74.2 (#16)
| Benchmark | GPT-5.1 | Kimi K3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.6% | 97.2% |
| LMArena Math | 1447 | 1491 |
| FrontierMath (Tiers 1-3) | — | 72.2% |
| FrontierMath Tier 4 | — | 39% |
| MathArena Final-Answer Competitions | — | 87.8% |
| ProofBench | — | 87% |
| Omni-MATH | 46.4% | — |
| LiveBench Math | 94.5% | — |
| FrontierMath (Feb 2025 set) | 31% | — |
| FrontierMath Tier 4 (v1) | 12.5% | — |
Knowledge Kimi K3 leads
GPT-5.1: 50.6 (#71), Kimi K3: 63.2 (#21)
| Benchmark | GPT-5.1 | Kimi K3 |
|---|---|---|
| GPQA Diamond | 87.6% | 93.1% |
| SimpleQA Verified | 48% | 50.6% |
| LMArena Expert | 1470 | 1521 |
| Humanity's Last Exam | 23.7% | — |
| MMLU-Pro | 57.9% | — |
| Vectara Hallucination Rate | 10.9% | — |
| GPQA (HELM) | 44.2% | — |
Multimodal GPT-5.1 leads
GPT-5.1: 44.8 (#19), Kimi K3: 37.8 (#70)
| Benchmark | GPT-5.1 | Kimi K3 |
|---|---|---|
| LMArena Vision | 1250 | — |
| VPCT | 58.7% | — |
| Blueprint-Bench 2 | — | 29.5% |
| Furniture Assembly | — | 34.2% |
| LMArena Document | 1403 | — |
Multilingual Kimi K3 leads
GPT-5.1: 53.8 (#56), Kimi K3: 56.3 (#21)
| Benchmark | GPT-5.1 | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1431 | 1466 |
| LMArena Chinese | 1495 | 1529 |
| LMArena French | 1450 | 1491 |
| LMArena German | 1438 | 1488 |
| LMArena Japanese | 1453 | 1487 |
| LMArena Korean | 1401 | 1458 |
| LMArena Russian | 1435 | 1482 |
| LMArena Spanish | 1433 | 1472 |
Instruction Following GPT-5.1 leads
GPT-5.1: 83.9 (#1), Kimi K3: 77.7 (#14)
| Benchmark | GPT-5.1 | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1443 | 1483 |
| LiveBench Instruction Following | 93.3% | — |
| IFEval | 93.5% | — |
Long Context GPT-5.1 leads
GPT-5.1: 47.6 (#14), Kimi K3: 45.8 (#29)
| Benchmark | GPT-5.1 | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1447 | 1494 |
| CL-bench | 23.7% | — |
| CL-bench Life | 17.3% | — |
Writing & Preference Kimi K3 leads
GPT-5.1: 64.5 (#55), Kimi K3: 76.6 (#4)
| Benchmark | GPT-5.1 | Kimi K3 |
|---|---|---|
| LMArena Text | 1443 | 1476 |
| LMArena Creative Writing | 1427 | 1454 |
| LMArena Multi-Turn | 1450 | 1488 |
| EQ-Bench Creative Writing | — | 2082 |
| WildBench | 86.3% | — |
| EQ-Bench 4 | — | 1339 |
| LiveBench Language | 80.2% | — |
Frequently asked questions
Is GPT-5.1 better than Kimi K3?
Kimi K3 is the stronger model overall, scoring 59.5 to 49.0 on the Noometry Index. GPT-5.1 costs 1.7× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Which is cheaper, GPT-5.1 or Kimi K3?
GPT-5.1 is cheaper. It lists at $1.25 per million input tokens and $10 per million output tokens; Kimi K3 lists at $3 and $15.
Is GPT-5.1 or Kimi K3 better for coding?
Kimi K3 scores higher on coding benchmarks: 61.0 versus 46.4 in the Noometry coding category.
Which has the bigger context window?
Kimi K3 does, with 1.05M tokens against 400K.
How many benchmarks do GPT-5.1 and Kimi K3 share?
35 benchmarks have published results for both models. GPT-5.1 has 63 scored results on Noometry and Kimi K3 has 53.