Model comparison
Gemini 3.1 Flash Lite vs Kimi K2.5
Kimi K2.5 is the stronger model overall, scoring 47.8 to 40.4 on the Noometry Index. Gemini 3.1 Flash Lite costs 1.6× less per token, which makes it the better buy when Kimi K2.5's lead doesn't matter for your workload.
Last verified . 33 shared benchmarks.
GoogleSummary
- They share 33 benchmarks with published results for both. Gemini 3.1 Flash Lite scores higher in 0 categories and Kimi K2.5 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Kimi K2.5 leads 53.6 to 41.7.
- The biggest single-benchmark swing is NYT Connections (extended): 8.2% for Gemini 3.1 Flash Lite and 69.9% for Kimi K2.5.
- Gemini 3.1 Flash Lite is cheaper at $0.25 / $1.50 per million input/output tokens, against $0.45 / $2.25 for Kimi K2.5.
- Gemini 3.1 Flash Lite accepts more context: 1.05M tokens versus 262K.
- Kimi K2.5 has downloadable open weights; the other is API-only.
Side by side
| Gemini 3.1 Flash Lite | Kimi K2.5 | |
|---|---|---|
| Provider | Moonshot AI | |
| Noometry Index | 40.4 | 47.8 |
| Released | 2026-03-03 | 2026-01-27 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 262K |
| Max output | 66K | 262K |
| Input $ / M tokens | $0.25 | $0.45 |
| Output $ / M tokens | $1.50 | $2.25 |
| Results tracked | 38 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.5 leads
Gemini 3.1 Flash Lite: 37.5 (#179), Kimi K2.5: 48.7 (#48)
| Benchmark | Gemini 3.1 Flash Lite | Kimi K2.5 |
|---|---|---|
| LMArena WebDev | 1256 | 1436 |
| SciCode | 41.9% | 49% |
| WeirdML | 52.2% | 45.6% |
| LMArena Coding | 1400 | 1474 |
| ALE-Bench | 797.73 | 821.65 |
| SWE-bench Verified | — | 73.8% |
| SWE-bench Verified (bash only) | — | 70.8% |
| SWE-bench Multilingual | — | 67.3% |
Agentic & Tool Use Kimi K2.5 leads
Gemini 3.1 Flash Lite: 29.9 (#72), Kimi K2.5: 34.0 (#44)
| Benchmark | Gemini 3.1 Flash Lite | Kimi K2.5 |
|---|---|---|
| Terminal-Bench | — | 43.2% |
| DeepResearch Bench | 37.3% | — |
| OSWorld | — | 63.3% |
| Vending-Bench 2 | — | 1,198 |
Reasoning Kimi K2.5 leads
Gemini 3.1 Flash Lite: 22.3 (#181), Kimi K2.5: 31.2 (#74)
| Benchmark | Gemini 3.1 Flash Lite | Kimi K2.5 |
|---|---|---|
| Kagi LLM Benchmark | 67.2% | 78.5% |
| NYT Connections (extended) | 8.2% | 69.9% |
| CritPt | 1.1% | 3.1% |
| Chess Puzzles | 25% | 12% |
| EnigmaEval | 3% | 3.4% |
| Thematic Generalization | 63.3% | 69.4% |
| LMArena Hard Prompts | 1407 | 1454 |
| Epoch Capabilities Index | 144.45 | 148.05 |
| ARC-AGI-2 | — | 11.8% |
| SimpleBench | — | 46.8% |
| ARC-AGI-1 | — | 65.3% |
| DTBench | 76.8% | — |
| LMCA | 35% | — |
| ForecastBench | 54.4 | — |
Math Kimi K2.5 leads
Gemini 3.1 Flash Lite: 39.5 (#82), Kimi K2.5: 50.3 (#46)
| Benchmark | Gemini 3.1 Flash Lite | Kimi K2.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 92.2% |
| LMArena Math | 1427 | 1470 |
| FrontierMath (Tiers 1-3) | 27.7% | — |
| MathArena Final-Answer Competitions | — | 62.3% |
| FrontierMath (Feb 2025 set) | — | 27.9% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge Kimi K2.5 leads
Gemini 3.1 Flash Lite: 41.7 (#95), Kimi K2.5: 53.6 (#52)
| Benchmark | Gemini 3.1 Flash Lite | Kimi K2.5 |
|---|---|---|
| GPQA Diamond | 81.8% | 87.6% |
| Humanity's Last Exam | 8.6% | 24.4% |
| Vectara Hallucination Rate | 8.2% | 14.2% |
| LMArena Expert | 1397 | 1466 |
| SimpleQA Verified | — | 34.3% |
Multimodal Kimi K2.5 leads
Gemini 3.1 Flash Lite: 38.7 (#57), Kimi K2.5: 40.4 (#37)
| Benchmark | Gemini 3.1 Flash Lite | Kimi K2.5 |
|---|---|---|
| LMArena Vision | 1240 | 1269 |
| LMArena Document | — | 1430 |
Multilingual Kimi K2.5 leads
Gemini 3.1 Flash Lite: 52.5 (#82), Kimi K2.5: 54.1 (#50)
| Benchmark | Gemini 3.1 Flash Lite | Kimi K2.5 |
|---|---|---|
| LMArena Non-English | 1410 | 1433 |
| LMArena Chinese | 1461 | 1495 |
| LMArena French | 1423 | 1454 |
| LMArena German | 1429 | 1441 |
| LMArena Japanese | 1411 | 1422 |
| LMArena Korean | 1393 | 1412 |
| LMArena Russian | 1420 | 1436 |
| LMArena Spanish | 1420 | 1450 |
Instruction Following Kimi K2.5 leads
Gemini 3.1 Flash Lite: 72.9 (#126), Kimi K2.5: 75.5 (#60)
| Benchmark | Gemini 3.1 Flash Lite | Kimi K2.5 |
|---|---|---|
| LMArena Instruction Following | 1376 | 1431 |
Long Context Kimi K2.5 leads
Gemini 3.1 Flash Lite: 42.5 (#117), Kimi K2.5: 52.1 (#7)
| Benchmark | Gemini 3.1 Flash Lite | Kimi K2.5 |
|---|---|---|
| LMArena Longer Query | 1394 | 1445 |
| Fiction.LiveBench | — | 86.1% |
| CL-bench | — | 19.3% |
| CL-bench Life | — | 13.2% |
Writing & Preference Kimi K2.5 leads
Gemini 3.1 Flash Lite: 61.0 (#90), Kimi K2.5: 65.3 (#51)
| Benchmark | Gemini 3.1 Flash Lite | Kimi K2.5 |
|---|---|---|
| LMArena Text | 1416 | 1445 |
| LMArena Creative Writing | 1401 | 1423 |
| LMArena Multi-Turn | 1417 | 1444 |
| EQ-Bench Creative Writing | — | 1579 |
Frequently asked questions
Is Gemini 3.1 Flash Lite better than Kimi K2.5?
Kimi K2.5 is the stronger model overall, scoring 47.8 to 40.4 on the Noometry Index. Gemini 3.1 Flash Lite costs 1.6× less per token, which makes it the better buy when Kimi K2.5's lead doesn't matter for your workload.
Which is cheaper, Gemini 3.1 Flash Lite or Kimi K2.5?
Gemini 3.1 Flash Lite is cheaper. It lists at $0.25 per million input tokens and $1.50 per million output tokens; Kimi K2.5 lists at $0.45 and $2.25.
Is Gemini 3.1 Flash Lite or Kimi K2.5 better for coding?
Kimi K2.5 scores higher on coding benchmarks: 48.7 versus 37.5 in the Noometry coding category.
Which has the bigger context window?
Gemini 3.1 Flash Lite does, with 1.05M tokens against 262K.
How many benchmarks do Gemini 3.1 Flash Lite and Kimi K2.5 share?
33 benchmarks have published results for both models. Gemini 3.1 Flash Lite has 38 scored results on Noometry and Kimi K2.5 has 51.