Model comparison
Gemini 3.8 Flash vs Kimi K2 (Jul 2025)
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 41.2 on the Noometry Index.
Last verified . 21 shared benchmarks.
GoogleSummary
- They share 21 benchmarks with published results for both. Gemini 3.8 Flash scores higher in 9 categories and Kimi K2 (Jul 2025) in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.8 Flash leads 76.9 to 23.3.
- The biggest single-benchmark swing is WeirdML: 84.8% for Gemini 3.8 Flash and 42.8% for Kimi K2 (Jul 2025).
- Kimi K2 (Jul 2025) is cheaper at $0.57 / $2.30 per million input/output tokens, against $0.75 / $3.75 for Gemini 3.8 Flash.
- Gemini 3.8 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 3.8 Flash | Kimi K2 (Jul 2025) | |
|---|---|---|
| Provider | Moonshot AI | |
| Noometry Index | 61.8 | 41.2 |
| Released | 2026-09-02 | 2025-07-12 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 262K |
| Max output | 66K | 262K |
| Input $ / M tokens | $0.75 | $0.57 |
| Output $ / M tokens | $3.75 | $2.30 |
| Results tracked | 50 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3.8 Flash leads
Gemini 3.8 Flash: 59.2 (#15), Kimi K2 (Jul 2025): 42.4 (#102)
| Benchmark | Gemini 3.8 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| WeirdML | 84.8% | 42.8% |
| LMArena Coding | 1510 | 1399 |
| ALE-Bench | 1,270 | 597.5 |
| DeepSWE | 73.8% | — |
| FrontierCode | 41.2% | — |
| SWE-bench Verified (bash only) | — | 63.4% |
| Aider Polyglot | — | 59.1% |
| CursorBench | 39.6% | — |
| LMArena WebDev | 1584 | — |
| FrontierSWE | 19.6% | — |
| SciCode | 56.6% | — |
| GSO | — | 4.9% |
Agentic & Tool Use Gemini 3.8 Flash leads
Gemini 3.8 Flash: 41.8 (#21), Kimi K2 (Jul 2025): 32.4 (#64)
| Benchmark | Gemini 3.8 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| Terminal-Bench | — | 35.7% |
| APEX-Agents | 64.3% | — |
| Berkeley Function Calling Leaderboard | — | 59.1% |
| Remote Labor Index | 5.8% | — |
| GDP.pdf | 23.4% | — |
| METR Time Horizons | — | 59.2% |
| Vending-Bench 2 | 5,094 | — |
Reasoning Gemini 3.8 Flash leads
Gemini 3.8 Flash: 76.9 (#5), Kimi K2 (Jul 2025): 23.3 (#179)
| Benchmark | Gemini 3.8 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Hard Prompts | 1508 | 1384 |
| Epoch Capabilities Index | 156.71 | 146.01 |
| ARC-AGI-2 | 89.2% | — |
| SimpleBench | — | 26.3% |
| Kagi LLM Benchmark | — | 64.4% |
| NYT Connections (extended) | 97.4% | — |
| ARC-AGI-1 | 98.5% | — |
| CritPt | 18.3% | — |
| Chess Puzzles | 61% | — |
| Mystery Game Puzzles | 47% | — |
| DTBench | 95.7% | — |
| LMCA | 52.9% | — |
| Surface Evolver Bench | 76.9% | — |
| ForecastBench | — | 60.2 |
Math Gemini 3.8 Flash leads
Gemini 3.8 Flash: 65.3 (#28), Kimi K2 (Jul 2025): 42.7 (#83)
| Benchmark | Gemini 3.8 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Math | 1528 | 1397 |
| FrontierMath (Tiers 1-3) | 68.4% | — |
| FrontierMath Tier 4 | 22% | — |
| OTIS Mock AIME 2024-2025 | 98.9% | — |
| ProofBench | 48% | — |
| Omni-MATH | — | 65.4% |
| FrontierMath (Feb 2025 set) | — | 21.4% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Gemini 3.8 Flash leads
Gemini 3.8 Flash: 74.8 (#2), Kimi K2 (Jul 2025): 37.3 (#157)
| Benchmark | Gemini 3.8 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Expert | 1524 | 1365 |
| GPQA Diamond | 95.4% | — |
| Humanity's Last Exam | 44.5% | — |
| SimpleQA Verified | 69.7% | — |
| MMLU-Pro | — | 81.9% |
| Confabulations | — | 20.4% |
| Vectara Hallucination Rate | — | 17.9% |
| GPQA (HELM) | — | 65.3% |
Multimodal Not comparable
Gemini 3.8 Flash: 40.7 (#45), Kimi K2 (Jul 2025): —
| Benchmark | Gemini 3.8 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Vision | 1314 | — |
| Blueprint-Bench 2 | 38.6% | — |
| Furniture Assembly | 31.7% | — |
Multilingual Gemini 3.8 Flash leads
Gemini 3.8 Flash: 58.0 (#5), Kimi K2 (Jul 2025): 49.6 (#130)
| Benchmark | Gemini 3.8 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1491 | 1372 |
| LMArena Chinese | 1554 | 1415 |
| LMArena French | 1498 | 1379 |
| LMArena German | 1493 | 1387 |
| LMArena Japanese | 1502 | 1349 |
| LMArena Korean | 1459 | 1325 |
| LMArena Russian | 1515 | 1385 |
| LMArena Spanish | 1485 | 1386 |
Instruction Following Gemini 3.8 Flash leads
Gemini 3.8 Flash: 78.0 (#13), Kimi K2 (Jul 2025): 71.1 (#156)
| Benchmark | Gemini 3.8 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Instruction Following | 1490 | 1348 |
| IFEval | — | 85% |
Long Context Gemini 3.8 Flash leads
Gemini 3.8 Flash: 46.3 (#24), Kimi K2 (Jul 2025): 41.2 (#145)
| Benchmark | Gemini 3.8 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Longer Query | 1508 | 1353 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 17.6% |
Writing & Preference Gemini 3.8 Flash leads
Gemini 3.8 Flash: 72.2 (#15), Kimi K2 (Jul 2025): 62.3 (#78)
| Benchmark | Gemini 3.8 Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1499 | 1380 |
| LMArena Creative Writing | 1492 | 1350 |
| EQ-Bench Creative Writing | 1748 | 1666 |
| LMArena Multi-Turn | 1501 | 1371 |
| Short-Story Creative Writing | — | 85.6% |
| WildBench | — | 86.2% |
Frequently asked questions
Is Gemini 3.8 Flash better than Kimi K2 (Jul 2025)?
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 41.2 on the Noometry Index.
Which is cheaper, Gemini 3.8 Flash 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; Gemini 3.8 Flash lists at $0.75 and $3.75.
Is Gemini 3.8 Flash or Kimi K2 (Jul 2025) better for coding?
Gemini 3.8 Flash scores higher on coding benchmarks: 59.2 versus 42.4 in the Noometry coding category.
Which has the bigger context window?
Gemini 3.8 Flash does, with 1.05M tokens against 262K.
How many benchmarks do Gemini 3.8 Flash and Kimi K2 (Jul 2025) share?
21 benchmarks have published results for both models. Gemini 3.8 Flash has 50 scored results on Noometry and Kimi K2 (Jul 2025) has 42.