Model comparison
GPT-5 Nano vs Kimi K2 (Jul 2025)
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 33.5 on the Noometry Index. GPT-5 Nano costs 7.3× less per token, which makes it the better buy when Kimi K2 (Jul 2025)'s lead doesn't matter for your workload.
Last verified . 34 shared benchmarks.
Summary
- They share 34 benchmarks with published results for both. GPT-5 Nano scores higher in 1 category and Kimi K2 (Jul 2025) in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Kimi K2 (Jul 2025) leads 62.3 to 39.1.
- The biggest single-benchmark swing is SWE-bench Verified (bash only): 34.8% for GPT-5 Nano and 63.4% for Kimi K2 (Jul 2025).
- GPT-5 Nano is cheaper at $0.05 / $0.40 per million input/output tokens, against $0.57 / $2.30 for Kimi K2 (Jul 2025).
- GPT-5 Nano accepts more context: 400K tokens versus 262K.
- Kimi K2 (Jul 2025) has downloadable open weights; the other is API-only.
Side by side
| GPT-5 Nano | Kimi K2 (Jul 2025) | |
|---|---|---|
| Provider | OpenAI | Moonshot AI |
| Noometry Index | 33.5 | 41.2 |
| Released | 2025-08-07 | 2025-07-12 |
| Weights | Proprietary | Open |
| Context window | 400K | 262K |
| Max output | 128K | 262K |
| Input $ / M tokens | $0.05 | $0.57 |
| Output $ / M tokens | $0.40 | $2.30 |
| Results tracked | 49 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2 (Jul 2025) leads
GPT-5 Nano: 33.6 (#254), Kimi K2 (Jul 2025): 42.4 (#102)
| Benchmark | GPT-5 Nano | Kimi K2 (Jul 2025) |
|---|---|---|
| SWE-bench Verified (bash only) | 34.8% | 63.4% |
| WeirdML | 38.1% | 42.8% |
| LMArena Coding | 1351 | 1399 |
| ALE-Bench | 718.67 | 597.5 |
| Aider Polyglot | — | 59.1% |
| GSO | — | 4.9% |
Agentic & Tool Use Kimi K2 (Jul 2025) leads
GPT-5 Nano: 25.8 (#106), Kimi K2 (Jul 2025): 32.4 (#64)
| Benchmark | GPT-5 Nano | Kimi K2 (Jul 2025) |
|---|---|---|
| Terminal-Bench | 21.8% | 35.7% |
| Berkeley Function Calling Leaderboard | 51.5% | 59.1% |
| METR Time Horizons | — | 59.2% |
Reasoning Kimi K2 (Jul 2025) leads
GPT-5 Nano: 16.3 (#306), Kimi K2 (Jul 2025): 23.3 (#179)
| Benchmark | GPT-5 Nano | Kimi K2 (Jul 2025) |
|---|---|---|
| Kagi LLM Benchmark | 62.2% | 64.4% |
| LMArena Hard Prompts | 1328 | 1384 |
| Epoch Capabilities Index | 139.38 | 146.01 |
| ForecastBench | 59.1 | 60.2 |
| ARC-AGI-2 | 2.6% | — |
| SimpleBench | — | 26.3% |
| ARC-AGI-1 | 20.7% | — |
| Chess Puzzles | 27% | — |
| Mystery Game Puzzles | 9% | — |
| DTBench | 62.7% | — |
| LMCA | 7.9% | — |
Math Kimi K2 (Jul 2025) leads
GPT-5 Nano: 29.4 (#241), Kimi K2 (Jul 2025): 42.7 (#83)
| Benchmark | GPT-5 Nano | Kimi K2 (Jul 2025) |
|---|---|---|
| Omni-MATH | 54.6% | 65.4% |
| LMArena Math | 1317 | 1397 |
| FrontierMath (Feb 2025 set) | 8.3% | 21.4% |
| FrontierMath Tier 4 (v1) | 2.1% | 0% |
| FrontierMath (Tiers 1-3) | 20% | — |
| FrontierMath Tier 4 | 2.4% | — |
| OTIS Mock AIME 2024-2025 | 81.1% | — |
| ProofBench | 12% | — |
| MATH Level 5 | 95.2% | — |
Knowledge Kimi K2 (Jul 2025) leads
GPT-5 Nano: 35.9 (#178), Kimi K2 (Jul 2025): 37.3 (#157)
| Benchmark | GPT-5 Nano | Kimi K2 (Jul 2025) |
|---|---|---|
| MMLU-Pro | 77.8% | 81.9% |
| Vectara Hallucination Rate | 10.5% | 17.9% |
| GPQA (HELM) | 67.9% | 65.3% |
| LMArena Expert | 1321 | 1365 |
| GPQA Diamond | 69.4% | — |
| SimpleQA Verified | 11.7% | — |
| Confabulations | — | 20.4% |
Multimodal Not comparable
GPT-5 Nano: 31.3 (#108), Kimi K2 (Jul 2025): —
| Benchmark | GPT-5 Nano | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Vision | 1159 | — |
| VPCT | 37.2% | — |
Multilingual Kimi K2 (Jul 2025) leads
GPT-5 Nano: 45.3 (#172), Kimi K2 (Jul 2025): 49.6 (#130)
| Benchmark | GPT-5 Nano | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1313 | 1372 |
| LMArena Chinese | 1356 | 1415 |
| LMArena German | 1327 | 1387 |
| LMArena Japanese | 1226 | 1349 |
| LMArena Korean | 1269 | 1325 |
| LMArena Russian | 1296 | 1385 |
| LMArena Spanish | 1360 | 1386 |
| LMArena French | — | 1379 |
Instruction Following GPT-5 Nano leads
GPT-5 Nano: 75.0 (#79), Kimi K2 (Jul 2025): 71.1 (#156)
| Benchmark | GPT-5 Nano | Kimi K2 (Jul 2025) |
|---|---|---|
| IFEval | 93.2% | 85% |
| LMArena Instruction Following | 1306 | 1348 |
Long Context Kimi K2 (Jul 2025) leads
GPT-5 Nano: 31.3 (#281), Kimi K2 (Jul 2025): 41.2 (#145)
| Benchmark | GPT-5 Nano | Kimi K2 (Jul 2025) |
|---|---|---|
| Fiction.LiveBench | 44.4% | 66.7% |
| LMArena Longer Query | 1312 | 1353 |
| CL-bench | — | 17.6% |
Writing & Preference Kimi K2 (Jul 2025) leads
GPT-5 Nano: 39.1 (#249), Kimi K2 (Jul 2025): 62.3 (#78)
| Benchmark | GPT-5 Nano | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1320 | 1380 |
| LMArena Creative Writing | 1249 | 1350 |
| EQ-Bench Creative Writing | 705 | 1666 |
| WildBench | 80.6% | 86.2% |
| LMArena Multi-Turn | 1311 | 1371 |
| Short-Story Creative Writing | — | 85.6% |
Frequently asked questions
Is GPT-5 Nano better than Kimi K2 (Jul 2025)?
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 33.5 on the Noometry Index. GPT-5 Nano costs 7.3× less per token, which makes it the better buy when Kimi K2 (Jul 2025)'s lead doesn't matter for your workload.
Which is cheaper, GPT-5 Nano or Kimi K2 (Jul 2025)?
GPT-5 Nano is cheaper. It lists at $0.05 per million input tokens and $0.40 per million output tokens; Kimi K2 (Jul 2025) lists at $0.57 and $2.30.
Is GPT-5 Nano or Kimi K2 (Jul 2025) better for coding?
Kimi K2 (Jul 2025) scores higher on coding benchmarks: 42.4 versus 33.6 in the Noometry coding category.
Which has the bigger context window?
GPT-5 Nano does, with 400K tokens against 262K.
How many benchmarks do GPT-5 Nano and Kimi K2 (Jul 2025) share?
34 benchmarks have published results for both models. GPT-5 Nano has 49 scored results on Noometry and Kimi K2 (Jul 2025) has 42.