Model comparison
GPT-4o mini vs Kimi K2.6
Kimi K2.6 is the stronger model overall, scoring 47.7 to 25.5 on the Noometry Index. GPT-4o mini costs 6.5× less per token, which makes it the better buy when Kimi K2.6's lead doesn't matter for your workload.
Last verified . 29 shared benchmarks.
Summary
- They share 29 benchmarks with published results for both. GPT-4o mini scores higher in 1 category and Kimi K2.6 in 9 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.6 leads 57.0 to 10.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.9% for GPT-4o mini and 96.1% for Kimi K2.6.
- GPT-4o mini is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.95 / $4 for Kimi K2.6.
- Kimi K2.6 accepts more context: 262K tokens versus 128K.
- Kimi K2.6 has downloadable open weights; the other is API-only.
Side by side
| GPT-4o mini | Kimi K2.6 | |
|---|---|---|
| Provider | OpenAI | Moonshot AI |
| Noometry Index | 25.5 | 47.7 |
| Released | 2024-07-18 | 2026-04-20 |
| Weights | Proprietary | Open |
| Context window | 128K | 262K |
| Max output | 16K | 262K |
| Input $ / M tokens | $0.15 | $0.95 |
| Output $ / M tokens | $0.60 | $4 |
| Results tracked | 60 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.6 leads
GPT-4o mini: 22.0 (#335), Kimi K2.6: 50.7 (#43)
| Benchmark | GPT-4o mini | Kimi K2.6 |
|---|---|---|
| WeirdML | 11.8% | 55.9% |
| LMArena Coding | 1290 | 1488 |
| SWE-bench Verified | — | 76.7% |
| Aider Polyglot | 3.6% | — |
| LMArena WebDev | — | 1509 |
| SciCode | — | 53.5% |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 43.1% | — |
| BigCodeBench Complete | 57.4% | — |
| ALE-Bench | — | 1,093 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 72.2% | — |
Agentic & Tool Use GPT-4o mini leads
GPT-4o mini: 27.5 (#101), Kimi K2.6: 21.9 (#137)
| Benchmark | GPT-4o mini | Kimi K2.6 |
|---|---|---|
| OSWorld 2.0 | — | 4.6% |
| BALROG | 17.4% | — |
| ExploitBench | — | 18.4% |
| GBAEval | — | 0.9% |
| GDP.pdf | — | 12% |
| Vending-Bench 2 | — | 6,205 |
Reasoning Kimi K2.6 leads
GPT-4o mini: 8.7 (#347), Kimi K2.6: 40.5 (#55)
| Benchmark | GPT-4o mini | Kimi K2.6 |
|---|---|---|
| Chess Puzzles | 0% | 26% |
| LMArena Hard Prompts | 1267 | 1470 |
| Mystery Game Puzzles | 12% | 18% |
| DTBench | 54.4% | 90.9% |
| LMCA | 10.4% | 37.3% |
| Epoch Capabilities Index | 126.56 | 151.05 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 10.7% | — |
| Kagi LLM Benchmark | 28.8% | — |
| NYT Connections (extended) | — | 87.2% |
| CritPt | — | 8% |
| EBR-Bench | — | 2.4% |
| LiveBench Reasoning | 32.8% | — |
| LiveBench Data Analysis | 50% | — |
| LiveBench | 41.3% | — |
| PIQA | 88.7% | — |
Math Kimi K2.6 leads
GPT-4o mini: 10.4 (#314), Kimi K2.6: 57.0 (#41)
| Benchmark | GPT-4o mini | Kimi K2.6 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 0.7% | 57.2% |
| OTIS Mock AIME 2024-2025 | 6.9% | 96.1% |
| LMArena Math | 1267 | 1475 |
| FrontierMath Tier 4 | — | 25.6% |
| MathArena Final-Answer Competitions | — | 72.9% |
| ProofBench | — | 16% |
| Omni-MATH | 28% | — |
| LiveBench Math | 36.3% | — |
| MATH Level 5 | 52.6% | — |
| FrontierMath (Feb 2025 set) | — | 39% |
| FrontierMath Tier 4 (v1) | — | 14.6% |
| GSM8K | 91.3% | — |
Knowledge Kimi K2.6 leads
GPT-4o mini: 17.7 (#284), Kimi K2.6: 54.0 (#54)
| Benchmark | GPT-4o mini | Kimi K2.6 |
|---|---|---|
| GPQA Diamond | 37.7% | 90.8% |
| SimpleQA Verified | 8.3% | 34.9% |
| LMArena Expert | 1235 | 1491 |
| MMLU-Pro | 60.3% | — |
| Confabulations | 37.2% | — |
| Vectara Hallucination Rate | — | 10.8% |
| GPQA (HELM) | 36.8% | — |
| BoolQ | 88.7% | — |
| MMLU | 81.8% | — |
Multimodal Kimi K2.6 leads
GPT-4o mini: 25.9 (#122), Kimi K2.6: 31.6 (#103)
| Benchmark | GPT-4o mini | Kimi K2.6 |
|---|---|---|
| LMArena Vision | 1066 | 1283 |
| Video-MME | 64.8% | — |
| GeoBench | 64% | — |
| VPCT | 34% | — |
| Blueprint-Bench 2 | — | 3.9% |
| Furniture Assembly | — | 21.7% |
| LMArena Document | — | 1451 |
Multilingual Kimi K2.6 leads
GPT-4o mini: 42.0 (#199), Kimi K2.6: 54.9 (#37)
| Benchmark | GPT-4o mini | Kimi K2.6 |
|---|---|---|
| LMArena Non-English | 1266 | 1446 |
| LMArena Chinese | 1265 | 1521 |
| LMArena French | 1297 | 1471 |
| LMArena German | 1272 | 1450 |
| LMArena Japanese | 1216 | 1443 |
| LMArena Korean | 1195 | 1427 |
| LMArena Russian | 1275 | 1446 |
| LMArena Spanish | 1276 | 1464 |
Instruction Following Kimi K2.6 leads
GPT-4o mini: 61.9 (#239), Kimi K2.6: 76.3 (#43)
| Benchmark | GPT-4o mini | Kimi K2.6 |
|---|---|---|
| LMArena Instruction Following | 1258 | 1451 |
| LiveBench Instruction Following | 56.8% | — |
| IFEval | 78.2% | — |
Long Context Kimi K2.6 leads
GPT-4o mini: 39.1 (#186), Kimi K2.6: 44.9 (#52)
| Benchmark | GPT-4o mini | Kimi K2.6 |
|---|---|---|
| LMArena Longer Query | 1289 | 1468 |
Writing & Preference Kimi K2.6 leads
GPT-4o mini: 39.5 (#248), Kimi K2.6: 68.5 (#26)
| Benchmark | GPT-4o mini | Kimi K2.6 |
|---|---|---|
| LMArena Text | 1286 | 1455 |
| LMArena Creative Writing | 1268 | 1434 |
| EQ-Bench Creative Writing | 873 | 1725 |
| LMArena Multi-Turn | 1285 | 1453 |
| Short-Story Creative Writing | 67.2% | — |
| WildBench | 79.1% | — |
| EQ-Bench 4 | — | 1202 |
| LiveBench Language | 28.6% | — |
Frequently asked questions
Is GPT-4o mini better than Kimi K2.6?
Kimi K2.6 is the stronger model overall, scoring 47.7 to 25.5 on the Noometry Index. GPT-4o mini costs 6.5× less per token, which makes it the better buy when Kimi K2.6's lead doesn't matter for your workload.
Which is cheaper, GPT-4o mini or Kimi K2.6?
GPT-4o mini is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Kimi K2.6 lists at $0.95 and $4.
Is GPT-4o mini or Kimi K2.6 better for coding?
Kimi K2.6 scores higher on coding benchmarks: 50.7 versus 22.0 in the Noometry coding category.
Which has the bigger context window?
Kimi K2.6 does, with 262K tokens against 128K.
How many benchmarks do GPT-4o mini and Kimi K2.6 share?
29 benchmarks have published results for both models. GPT-4o mini has 60 scored results on Noometry and Kimi K2.6 has 51.