Model comparison
Gemini 3.8 Flash vs Kimi K2.6
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 47.7 on the Noometry Index.
Last verified . 40 shared benchmarks.
GoogleSummary
- They share 40 benchmarks with published results for both. Gemini 3.8 Flash scores higher in 10 categories and Kimi K2.6 in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.8 Flash leads 76.9 to 40.5.
- The biggest single-benchmark swing is Chess Puzzles: 61% for Gemini 3.8 Flash and 26% for Kimi K2.6.
- Gemini 3.8 Flash is cheaper at $0.75 / $3.75 per million input/output tokens, against $0.95 / $4 for Kimi K2.6.
- Gemini 3.8 Flash accepts more context: 1.05M tokens versus 262K.
- Kimi K2.6 has downloadable open weights; the other is API-only.
Side by side
| Gemini 3.8 Flash | Kimi K2.6 | |
|---|---|---|
| Provider | Moonshot AI | |
| Noometry Index | 61.8 | 47.7 |
| Released | 2026-09-02 | 2026-04-20 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 262K |
| Max output | 66K | 262K |
| Input $ / M tokens | $0.75 | $0.95 |
| Output $ / M tokens | $3.75 | $4 |
| Results tracked | 50 | 51 |
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 K2.6: 50.7 (#43)
| Benchmark | Gemini 3.8 Flash | Kimi K2.6 |
|---|---|---|
| LMArena WebDev | 1584 | 1509 |
| SciCode | 56.6% | 53.5% |
| WeirdML | 84.8% | 55.9% |
| LMArena Coding | 1510 | 1488 |
| ALE-Bench | 1,270 | 1,093 |
| SWE-bench Verified | — | 76.7% |
| DeepSWE | 73.8% | — |
| FrontierCode | 41.2% | — |
| CursorBench | 39.6% | — |
| FrontierSWE | 19.6% | — |
Agentic & Tool Use Gemini 3.8 Flash leads
Gemini 3.8 Flash: 41.8 (#21), Kimi K2.6: 21.9 (#137)
| Benchmark | Gemini 3.8 Flash | Kimi K2.6 |
|---|---|---|
| GDP.pdf | 23.4% | 12% |
| Vending-Bench 2 | 5,094 | 6,205 |
| APEX-Agents | 64.3% | — |
| OSWorld 2.0 | — | 4.6% |
| Remote Labor Index | 5.8% | — |
| ExploitBench | — | 18.4% |
| GBAEval | — | 0.9% |
Reasoning Gemini 3.8 Flash leads
Gemini 3.8 Flash: 76.9 (#5), Kimi K2.6: 40.5 (#55)
| Benchmark | Gemini 3.8 Flash | Kimi K2.6 |
|---|---|---|
| NYT Connections (extended) | 97.4% | 87.2% |
| CritPt | 18.3% | 8% |
| Chess Puzzles | 61% | 26% |
| LMArena Hard Prompts | 1508 | 1470 |
| Mystery Game Puzzles | 47% | 18% |
| DTBench | 95.7% | 90.9% |
| LMCA | 52.9% | 37.3% |
| Epoch Capabilities Index | 156.71 | 151.05 |
| ARC-AGI-2 | 89.2% | — |
| ARC-AGI-1 | 98.5% | — |
| EBR-Bench | — | 2.4% |
| Surface Evolver Bench | 76.9% | — |
Math Gemini 3.8 Flash leads
Gemini 3.8 Flash: 65.3 (#28), Kimi K2.6: 57.0 (#41)
| Benchmark | Gemini 3.8 Flash | Kimi K2.6 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 68.4% | 57.2% |
| FrontierMath Tier 4 | 22% | 25.6% |
| OTIS Mock AIME 2024-2025 | 98.9% | 96.1% |
| ProofBench | 48% | 16% |
| LMArena Math | 1528 | 1475 |
| MathArena Final-Answer Competitions | — | 72.9% |
| FrontierMath (Feb 2025 set) | — | 39% |
| FrontierMath Tier 4 (v1) | — | 14.6% |
Knowledge Gemini 3.8 Flash leads
Gemini 3.8 Flash: 74.8 (#2), Kimi K2.6: 54.0 (#54)
| Benchmark | Gemini 3.8 Flash | Kimi K2.6 |
|---|---|---|
| GPQA Diamond | 95.4% | 90.8% |
| SimpleQA Verified | 69.7% | 34.9% |
| LMArena Expert | 1524 | 1491 |
| Humanity's Last Exam | 44.5% | — |
| Vectara Hallucination Rate | — | 10.8% |
Multimodal Gemini 3.8 Flash leads
Gemini 3.8 Flash: 40.7 (#45), Kimi K2.6: 31.6 (#103)
| Benchmark | Gemini 3.8 Flash | Kimi K2.6 |
|---|---|---|
| LMArena Vision | 1314 | 1283 |
| Blueprint-Bench 2 | 38.6% | 3.9% |
| Furniture Assembly | 31.7% | 21.7% |
| LMArena Document | — | 1451 |
Multilingual Gemini 3.8 Flash leads
Gemini 3.8 Flash: 58.0 (#5), Kimi K2.6: 54.9 (#37)
| Benchmark | Gemini 3.8 Flash | Kimi K2.6 |
|---|---|---|
| LMArena Non-English | 1491 | 1446 |
| LMArena Chinese | 1554 | 1521 |
| LMArena French | 1498 | 1471 |
| LMArena German | 1493 | 1450 |
| LMArena Japanese | 1502 | 1443 |
| LMArena Korean | 1459 | 1427 |
| LMArena Russian | 1515 | 1446 |
| LMArena Spanish | 1485 | 1464 |
Instruction Following Gemini 3.8 Flash leads
Gemini 3.8 Flash: 78.0 (#13), Kimi K2.6: 76.3 (#43)
| Benchmark | Gemini 3.8 Flash | Kimi K2.6 |
|---|---|---|
| LMArena Instruction Following | 1490 | 1451 |
Long Context Gemini 3.8 Flash leads
Gemini 3.8 Flash: 46.3 (#24), Kimi K2.6: 44.9 (#52)
| Benchmark | Gemini 3.8 Flash | Kimi K2.6 |
|---|---|---|
| LMArena Longer Query | 1508 | 1468 |
Writing & Preference Gemini 3.8 Flash leads
Gemini 3.8 Flash: 72.2 (#15), Kimi K2.6: 68.5 (#26)
| Benchmark | Gemini 3.8 Flash | Kimi K2.6 |
|---|---|---|
| LMArena Text | 1499 | 1455 |
| LMArena Creative Writing | 1492 | 1434 |
| EQ-Bench Creative Writing | 1748 | 1725 |
| LMArena Multi-Turn | 1501 | 1453 |
| EQ-Bench 4 | — | 1202 |
Frequently asked questions
Is Gemini 3.8 Flash better than Kimi K2.6?
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 47.7 on the Noometry Index.
Which is cheaper, Gemini 3.8 Flash or Kimi K2.6?
Gemini 3.8 Flash is cheaper. It lists at $0.75 per million input tokens and $3.75 per million output tokens; Kimi K2.6 lists at $0.95 and $4.
Is Gemini 3.8 Flash or Kimi K2.6 better for coding?
Gemini 3.8 Flash scores higher on coding benchmarks: 59.2 versus 50.7 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 K2.6 share?
40 benchmarks have published results for both models. Gemini 3.8 Flash has 50 scored results on Noometry and Kimi K2.6 has 51.