Model comparison
GPT-4o vs Kimi K2 (Jul 2025)
Kimi K2 (Jul 2025) is the stronger model overall, scoring 40.9 to 28.5 on the Noometry Index.
Last verified . 35 shared benchmarks.
Summary
- They share 35 benchmarks with published results for both. GPT-4o scores higher in 0 categories and Kimi K2 (Jul 2025) in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2 (Jul 2025) leads 40.1 to 10.2.
- The biggest single-benchmark swing is SWE-bench Verified (bash only): 21.6% for GPT-4o and 63.4% for Kimi K2 (Jul 2025).
- Kimi K2 (Jul 2025) is cheaper at $0.57 / $2.30 per million input/output tokens, against $2.50 / $10 for GPT-4o.
- Kimi K2 (Jul 2025) accepts more context: 262K tokens versus 128K.
- Kimi K2 (Jul 2025) has downloadable open weights; the other is API-only.
Side by side
| GPT-4o | Kimi K2 (Jul 2025) | |
|---|---|---|
| Provider | OpenAI | Moonshot AI |
| Noometry Index | 28.5 | 40.9 |
| Released | 2024-05-13 | 2025-07-12 |
| Weights | Proprietary | Open |
| Context window | 128K | 262K |
| Max output | 16K | 262K |
| Input $ / M tokens | $2.50 | $0.57 |
| Output $ / M tokens | $10 | $2.30 |
| Results tracked | 72 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2 (Jul 2025) leads
GPT-4o: 25.1 (#323), Kimi K2 (Jul 2025): 43.1 (#83)
| Benchmark | GPT-4o | Kimi K2 (Jul 2025) |
|---|---|---|
| SWE-bench Verified (bash only) | 21.6% | 63.4% |
| Aider Polyglot | 45.3% | 59.1% |
| GSO | 0% | 4.9% |
| WeirdML | 25.1% | 42.8% |
| LMArena Coding | 1297 | 1399 |
| SWE-bench Verified | 31% | — |
| BigCodeBench Instruct | 51.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 61.1% | — |
| CadEval | 26% | — |
| ALE-Bench | — | 597.5 |
| HumanEval+ | 87.2% | — |
| MBPP+ | 72.2% | — |
Agentic & Tool Use Kimi K2 (Jul 2025) leads
GPT-4o: 20.8 (#133), Kimi K2 (Jul 2025): 32.2 (#59)
| Benchmark | GPT-4o | Kimi K2 (Jul 2025) |
|---|---|---|
| METR Time Horizons | 40.8% | 59.2% |
| Terminal-Bench | — | 35.7% |
| Berkeley Function Calling Leaderboard | — | 59.1% |
| GDPval | 9.9% | — |
| TheAgentCompany | 8.6% | — |
| Cybench | 12.5% | — |
| BALROG | 32.3% | — |
| LMArena Search | 1006 | — |
Reasoning Kimi K2 (Jul 2025) leads
GPT-4o: 9.4 (#338), Kimi K2 (Jul 2025): 22.7 (#173)
| Benchmark | GPT-4o | Kimi K2 (Jul 2025) |
|---|---|---|
| SimpleBench | 17.8% | 26.3% |
| LMArena Hard Prompts | 1281 | 1384 |
| Epoch Capabilities Index | 128.97 | 146.02 |
| ForecastBench | 57.7 | 60.2 |
| ARC-AGI-2 | 0% | — |
| Kagi LLM Benchmark | — | 64.4% |
| ARC-AGI-1 | 4.5% | — |
| CritPt | 0% | — |
| Chess Puzzles | 13% | — |
| EnigmaEval | 0.8% | — |
| LiveBench Reasoning | 55.8% | — |
| DTBench | 64.5% | — |
| LiveBench Data Analysis | 60.9% | — |
| LMCA | 16.6% | — |
| LiveBench | 55.3% | — |
Math Kimi K2 (Jul 2025) leads
GPT-4o: 10.2 (#303), Kimi K2 (Jul 2025): 40.1 (#81)
| Benchmark | GPT-4o | Kimi K2 (Jul 2025) |
|---|---|---|
| Omni-MATH | 29.3% | 65.4% |
| LMArena Math | 1285 | 1396 |
| FrontierMath (Feb 2025 set) | 0.3% | 21.4% |
| FrontierMath (Tiers 1-3) | 0.4% | — |
| OTIS Mock AIME 2024-2025 | 6.4% | — |
| LiveBench Math | 49.5% | — |
| MATH Level 5 | 53.3% | — |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Kimi K2 (Jul 2025) leads
GPT-4o: 28.8 (#235), Kimi K2 (Jul 2025): 36.9 (#153)
| Benchmark | GPT-4o | Kimi K2 (Jul 2025) |
|---|---|---|
| MMLU-Pro | 71.3% | 81.9% |
| Confabulations | 15.3% | 20.4% |
| Vectara Hallucination Rate | 9.6% | 17.9% |
| GPQA (HELM) | 52% | 65.3% |
| LMArena Expert | 1250 | 1363 |
| GPQA Diamond | 49.2% | — |
| Humanity's Last Exam | 2.7% | — |
| SimpleQA Verified | 26% | — |
| MMLU | 88.1% | — |
Multimodal Not comparable
GPT-4o: 33.9 (#85), Kimi K2 (Jul 2025): —
| Benchmark | GPT-4o | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Vision | 1137 | — |
| Video-MME | 71.9% | — |
| GeoBench | 71% | — |
| VPCT | 40% | — |
| ScienceQA | 88.5% | — |
Multilingual Kimi K2 (Jul 2025) leads
GPT-4o: 43.4 (#181), Kimi K2 (Jul 2025): 49.8 (#125)
| Benchmark | GPT-4o | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1283 | 1372 |
| LMArena Chinese | 1277 | 1416 |
| LMArena French | 1303 | 1379 |
| LMArena German | 1282 | 1386 |
| LMArena Japanese | 1257 | 1349 |
| LMArena Korean | 1234 | 1324 |
| LMArena Russian | 1286 | 1384 |
| LMArena Spanish | 1292 | 1383 |
Instruction Following Kimi K2 (Jul 2025) leads
GPT-4o: 66.8 (#202), Kimi K2 (Jul 2025): 71.3 (#151)
| Benchmark | GPT-4o | Kimi K2 (Jul 2025) |
|---|---|---|
| IFEval | 81.7% | 85% |
| LMArena Instruction Following | 1278 | 1347 |
| LiveBench Instruction Following | 68.6% | — |
Long Context Kimi K2 (Jul 2025) leads
GPT-4o: 39.4 (#174), Kimi K2 (Jul 2025): 41.2 (#140)
| Benchmark | GPT-4o | Kimi K2 (Jul 2025) |
|---|---|---|
| Fiction.LiveBench | 66.7% | 66.7% |
| LMArena Longer Query | 1289 | 1353 |
| CL-bench | — | 17.6% |
Writing & Preference Kimi K2 (Jul 2025) leads
GPT-4o: 52.8 (#160), Kimi K2 (Jul 2025): 62.5 (#76)
| Benchmark | GPT-4o | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1300 | 1380 |
| LMArena Creative Writing | 1291 | 1349 |
| Short-Story Creative Writing | 81.8% | 85.6% |
| WildBench | 82.8% | 86.2% |
| LMArena Multi-Turn | 1302 | 1371 |
| EQ-Bench Creative Writing | — | 1666 |
| LiveBench Language | 47.6% | — |
Frequently asked questions
Is GPT-4o better than Kimi K2 (Jul 2025)?
Kimi K2 (Jul 2025) is the stronger model overall, scoring 40.9 to 28.5 on the Noometry Index.
Which is cheaper, GPT-4o 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; GPT-4o lists at $2.50 and $10.
Is GPT-4o or Kimi K2 (Jul 2025) better for coding?
Kimi K2 (Jul 2025) scores higher on coding benchmarks: 43.1 versus 25.1 in the Noometry coding category.
Which has the bigger context window?
Kimi K2 (Jul 2025) does, with 262K tokens against 128K.
How many benchmarks do GPT-4o and Kimi K2 (Jul 2025) share?
35 benchmarks have published results for both models. GPT-4o has 72 scored results on Noometry and Kimi K2 (Jul 2025) has 42.