Model comparison
Gemini 2.5 Flash vs Kimi K2 (Jul 2025)
Kimi K2 (Jul 2025) is the stronger model overall, scoring 40.9 to 39.0 on the Noometry Index.
Last verified . 39 shared benchmarks.
GoogleSummary
- They share 39 benchmarks with published results for both. Gemini 2.5 Flash scores higher in 3 categories and Kimi K2 (Jul 2025) in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Kimi K2 (Jul 2025) leads 62.5 to 54.0.
- The biggest single-benchmark swing is SWE-bench Verified (bash only): 28.7% for Gemini 2.5 Flash and 63.4% for Kimi K2 (Jul 2025).
- Gemini 2.5 Flash is cheaper at $0.30 / $2.50 per million input/output tokens, against $0.57 / $2.30 for Kimi K2 (Jul 2025).
- Gemini 2.5 Flash accepts more context: 1.05M tokens versus 262K.
- Kimi K2 (Jul 2025) has downloadable open weights; the other is API-only.
Side by side
| Gemini 2.5 Flash | Kimi K2 (Jul 2025) | |
|---|---|---|
| Provider | Moonshot AI | |
| Noometry Index | 39.0 | 40.9 |
| Released | 2025-04-17 | 2025-07-12 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 262K |
| Max output | 66K | 262K |
| Input $ / M tokens | $0.30 | $0.57 |
| Output $ / M tokens | $2.50 | $2.30 |
| Results tracked | 54 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2 (Jul 2025) leads
Gemini 2.5 Flash: 35.5 (#210), Kimi K2 (Jul 2025): 43.1 (#83)
| Benchmark | Gemini 2.5 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| SWE-bench Verified (bash only) | 28.7% | 63.4% |
| Aider Polyglot | 55.1% | 59.1% |
| WeirdML | 41.9% | 42.8% |
| LMArena Coding | 1424 | 1399 |
| ALE-Bench | 661.88 | 597.5 |
| GSO | — | 4.9% |
Agentic & Tool Use Kimi K2 (Jul 2025) leads
Gemini 2.5 Flash: 30.6 (#67), Kimi K2 (Jul 2025): 32.2 (#59)
| Benchmark | Gemini 2.5 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| Terminal-Bench | 17.1% | 35.7% |
| Berkeley Function Calling Leaderboard | 56.2% | 59.1% |
| TheAgentCompany | 41.1% | — |
| BALROG | 33.5% | — |
| METR Time Horizons | — | 59.2% |
| Vending-Bench 2 | 548.84 | — |
Reasoning Kimi K2 (Jul 2025) leads
Gemini 2.5 Flash: 18.2 (#271), Kimi K2 (Jul 2025): 22.7 (#173)
| Benchmark | Gemini 2.5 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| SimpleBench | 41.2% | 26.3% |
| Kagi LLM Benchmark | 56.8% | 64.4% |
| LMArena Hard Prompts | 1422 | 1384 |
| Epoch Capabilities Index | 143.04 | 146.02 |
| ForecastBench | 60.6 | 60.2 |
| ARC-AGI-2 | 2.5% | — |
| ARC-AGI-1 | 33.3% | — |
| CritPt | 1.1% | — |
| EnigmaEval | 2.7% | — |
| DTBench | 76.5% | — |
| LMCA | 27.5% | — |
Math Kimi K2 (Jul 2025) leads
Gemini 2.5 Flash: 38.1 (#87), Kimi K2 (Jul 2025): 40.1 (#81)
| Benchmark | Gemini 2.5 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| Omni-MATH | 38.5% | 65.4% |
| LMArena Math | 1415 | 1396 |
| FrontierMath (Feb 2025 set) | 4.8% | 21.4% |
| FrontierMath Tier 4 (v1) | 4.2% | 0% |
| OTIS Mock AIME 2024-2025 | 73.1% | — |
Knowledge Too close to call
Gemini 2.5 Flash: 36.1 (#162), Kimi K2 (Jul 2025): 36.9 (#153)
| Benchmark | Gemini 2.5 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| MMLU-Pro | 63.9% | 81.9% |
| Confabulations | 16.8% | 20.4% |
| Vectara Hallucination Rate | 7.8% | 17.9% |
| GPQA (HELM) | 39% | 65.3% |
| LMArena Expert | 1426 | 1363 |
| Humanity's Last Exam | 12.1% | — |
Multimodal Not comparable
Gemini 2.5 Flash: 41.2 (#31), Kimi K2 (Jul 2025): —
| Benchmark | Gemini 2.5 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Vision | 1254 | — |
| GeoBench | 76% | — |
| VPCT | 46.2% | — |
| SpatialViz-Bench | 36.9% | — |
Multilingual Gemini 2.5 Flash leads
Gemini 2.5 Flash: 52.5 (#83), Kimi K2 (Jul 2025): 49.8 (#125)
| Benchmark | Gemini 2.5 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1410 | 1372 |
| LMArena Chinese | 1450 | 1416 |
| LMArena French | 1433 | 1379 |
| LMArena German | 1419 | 1386 |
| LMArena Japanese | 1403 | 1349 |
| LMArena Korean | 1385 | 1324 |
| LMArena Russian | 1415 | 1384 |
| LMArena Spanish | 1421 | 1383 |
Instruction Following Gemini 2.5 Flash leads
Gemini 2.5 Flash: 75.9 (#50), Kimi K2 (Jul 2025): 71.3 (#151)
| Benchmark | Gemini 2.5 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| IFEval | 89.8% | 85% |
| LMArena Instruction Following | 1405 | 1347 |
Long Context Gemini 2.5 Flash leads
Gemini 2.5 Flash: 47.5 (#16), Kimi K2 (Jul 2025): 41.2 (#140)
| Benchmark | Gemini 2.5 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| Fiction.LiveBench | 77.8% | 66.7% |
| LMArena Longer Query | 1419 | 1353 |
| CL-bench | — | 17.6% |
Writing & Preference Kimi K2 (Jul 2025) leads
Gemini 2.5 Flash: 54.0 (#153), Kimi K2 (Jul 2025): 62.5 (#76)
| Benchmark | Gemini 2.5 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1417 | 1380 |
| LMArena Creative Writing | 1400 | 1349 |
| Short-Story Creative Writing | 76.5% | 85.6% |
| EQ-Bench Creative Writing | 1137 | 1666 |
| WildBench | 81.7% | 86.2% |
| LMArena Multi-Turn | 1408 | 1371 |
Frequently asked questions
Is Gemini 2.5 Flash better than Kimi K2 (Jul 2025)?
Kimi K2 (Jul 2025) is the stronger model overall, scoring 40.9 to 39.0 on the Noometry Index.
Which is cheaper, Gemini 2.5 Flash or Kimi K2 (Jul 2025)?
Gemini 2.5 Flash is cheaper. It lists at $0.30 per million input tokens and $2.50 per million output tokens; Kimi K2 (Jul 2025) lists at $0.57 and $2.30.
Is Gemini 2.5 Flash or Kimi K2 (Jul 2025) better for coding?
Kimi K2 (Jul 2025) scores higher on coding benchmarks: 43.1 versus 35.5 in the Noometry coding category.
Which has the bigger context window?
Gemini 2.5 Flash does, with 1.05M tokens against 262K.
How many benchmarks do Gemini 2.5 Flash and Kimi K2 (Jul 2025) share?
39 benchmarks have published results for both models. Gemini 2.5 Flash has 54 scored results on Noometry and Kimi K2 (Jul 2025) has 42.