Model comparison
Gemini 2.5 Flash-Lite vs Kimi K3
Kimi K3 is the stronger model overall, scoring 59.5 to 37.0 on the Noometry Index. Gemini 2.5 Flash-Lite costs 34× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Last verified . 22 shared benchmarks.
GoogleSummary
- They share 22 benchmarks with published results for both. Gemini 2.5 Flash-Lite scores higher in 0 categories and Kimi K3 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Kimi K3 leads 63.0 to 22.2.
- The biggest single-benchmark swing is WeirdML: 35.2% for Gemini 2.5 Flash-Lite and 82.6% for Kimi K3.
- Gemini 2.5 Flash-Lite is cheaper at $0.10 / $0.40 per million input/output tokens, against $3 / $15 for Kimi K3.
- Kimi K3 has downloadable open weights; the other is API-only.
Side by side
| Gemini 2.5 Flash-Lite | Kimi K3 | |
|---|---|---|
| Provider | Moonshot AI | |
| Noometry Index | 37.0 | 59.5 |
| Released | 2025-06-17 | 2026-07-16 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 1.05M |
| Max output | 66K | 1.05M |
| Input $ / M tokens | $0.10 | $3 |
| Output $ / M tokens | $0.40 | $15 |
| Results tracked | 33 | 53 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
Gemini 2.5 Flash-Lite: 38.5 (#173), Kimi K3: 61.0 (#10)
| Benchmark | Gemini 2.5 Flash-Lite | Kimi K3 |
|---|---|---|
| WeirdML | 35.2% | 82.6% |
| LMArena Coding | 1373 | 1508 |
| ALE-Bench | 325.9 | 1,524 |
| DeepSWE | — | 68.5% |
| FrontierCode | — | 44.2% |
| LMArena WebDev | — | 1654 |
| FrontierSWE | — | 25.9% |
| SciCode | — | 59.5% |
Agentic & Tool Use Kimi K3 leads
Gemini 2.5 Flash-Lite: 28.0 (#96), Kimi K3: 41.8 (#20)
| Benchmark | Gemini 2.5 Flash-Lite | Kimi K3 |
|---|---|---|
| APEX-Agents | — | 50.6% |
| Berkeley Function Calling Leaderboard | 36.9% | — |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| GDP.pdf | — | 19% |
| Vending-Bench 2 | — | 5,165 |
Reasoning Kimi K3 leads
Gemini 2.5 Flash-Lite: 22.2 (#205), Kimi K3: 63.0 (#17)
| Benchmark | Gemini 2.5 Flash-Lite | Kimi K3 |
|---|---|---|
| LMArena Hard Prompts | 1377 | 1496 |
| DTBench | 62.8% | 91.2% |
| LMCA | 18.1% | 52.7% |
| Epoch Capabilities Index | 133.94 | 157.45 |
| ARC-AGI-2 | — | 60.4% |
| SimpleBench | — | 60.7% |
| Kagi LLM Benchmark | 40.5% | — |
| NYT Connections (extended) | — | 93.6% |
| ARC-AGI-1 | — | 94.5% |
| CritPt | — | 23.4% |
| Chess Puzzles | — | 39% |
| Mystery Game Puzzles | — | 26% |
| Surface Evolver Bench | — | 95% |
| ForecastBench | — | 61.1 |
Math Kimi K3 leads
Gemini 2.5 Flash-Lite: 38.0 (#144), Kimi K3: 74.2 (#16)
| Benchmark | Gemini 2.5 Flash-Lite | Kimi K3 |
|---|---|---|
| LMArena Math | 1373 | 1491 |
| FrontierMath (Tiers 1-3) | — | 72.2% |
| FrontierMath Tier 4 | — | 39% |
| MathArena Final-Answer Competitions | — | 87.8% |
| OTIS Mock AIME 2024-2025 | — | 97.2% |
| ProofBench | — | 87% |
| Omni-MATH | 48% | — |
Knowledge Kimi K3 leads
Gemini 2.5 Flash-Lite: 32.5 (#210), Kimi K3: 63.2 (#21)
| Benchmark | Gemini 2.5 Flash-Lite | Kimi K3 |
|---|---|---|
| LMArena Expert | 1373 | 1521 |
| GPQA Diamond | — | 93.1% |
| SimpleQA Verified | — | 50.6% |
| MMLU-Pro | 53.7% | — |
| Vectara Hallucination Rate | 3.3% | — |
| GPQA (HELM) | 30.9% | — |
Multimodal Kimi K3 leads
Gemini 2.5 Flash-Lite: 29.1 (#114), Kimi K3: 37.8 (#70)
| Benchmark | Gemini 2.5 Flash-Lite | Kimi K3 |
|---|---|---|
| LMArena Vision | 1198 | — |
| VPCT | 30% | — |
| Blueprint-Bench 2 | — | 29.5% |
| Furniture Assembly | — | 34.2% |
Multilingual Kimi K3 leads
Gemini 2.5 Flash-Lite: 49.3 (#134), Kimi K3: 56.3 (#21)
| Benchmark | Gemini 2.5 Flash-Lite | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1369 | 1466 |
| LMArena Chinese | 1404 | 1529 |
| LMArena French | 1388 | 1491 |
| LMArena German | 1389 | 1488 |
| LMArena Japanese | 1359 | 1487 |
| LMArena Korean | 1360 | 1458 |
| LMArena Russian | 1373 | 1482 |
| LMArena Spanish | 1396 | 1472 |
Instruction Following Kimi K3 leads
Gemini 2.5 Flash-Lite: 70.0 (#168), Kimi K3: 77.7 (#14)
| Benchmark | Gemini 2.5 Flash-Lite | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1367 | 1483 |
| IFEval | 81% | — |
Long Context Kimi K3 leads
Gemini 2.5 Flash-Lite: 33.3 (#262), Kimi K3: 45.8 (#29)
| Benchmark | Gemini 2.5 Flash-Lite | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1373 | 1494 |
| Fiction.LiveBench | 47.2% | — |
Writing & Preference Kimi K3 leads
Gemini 2.5 Flash-Lite: 56.8 (#135), Kimi K3: 76.6 (#4)
| Benchmark | Gemini 2.5 Flash-Lite | Kimi K3 |
|---|---|---|
| LMArena Text | 1379 | 1476 |
| LMArena Creative Writing | 1367 | 1454 |
| LMArena Multi-Turn | 1366 | 1488 |
| EQ-Bench Creative Writing | — | 2082 |
| WildBench | 81.8% | — |
| EQ-Bench 4 | — | 1339 |
Frequently asked questions
Is Gemini 2.5 Flash-Lite better than Kimi K3?
Kimi K3 is the stronger model overall, scoring 59.5 to 37.0 on the Noometry Index. Gemini 2.5 Flash-Lite costs 34× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Which is cheaper, Gemini 2.5 Flash-Lite or Kimi K3?
Gemini 2.5 Flash-Lite is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; Kimi K3 lists at $3 and $15.
Is Gemini 2.5 Flash-Lite or Kimi K3 better for coding?
Kimi K3 scores higher on coding benchmarks: 61.0 versus 38.5 in the Noometry coding category.
Which has the bigger context window?
Both accept 1.05M tokens.
How many benchmarks do Gemini 2.5 Flash-Lite and Kimi K3 share?
22 benchmarks have published results for both models. Gemini 2.5 Flash-Lite has 33 scored results on Noometry and Kimi K3 has 53.