Up to four models
Gemini 3.5 Flash-Lite vs Gemini 3.1 Flash-Lite
On the index (GPT-4 = 100), Gemini 3.5 Flash-Lite scores 141 and Gemini 3.1 Flash-Lite 139. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
141125–156
WeightsClosed
ReleasedJul 2026
Price in / out$0.30 / $2.5
Context1.05M
Core tests taken5 of 9
139122–155
WeightsClosed
ReleasedMar 2026
Price in / out$0.25 / $1.5
Context1.05M
Core tests taken4 of 9
Gemini 3.5 Flash-Lite and Gemini 3.1 Flash-Lite are too close to call (2 points apart, within the uncertainty), ahead on 2 of 3 shared core tests, at 1.5× the price.
Test by testthird-party results
Every result as a table
| Test | Gemini 3.5 Flash-Lite | Gemini 3.1 Flash-Lite |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 83.3% | 81.8% |
| GPQA Diamond (Vals AI) | 83.8% | 81.1% |
| MMLU Pro | 85.8% | 86.2% |
| MMMU Pro | 83.6% | 82.5% |
| Maths | ||
| FrontierMath (tiers 1-3) | 26.0% | 27.7% |
| AIME-style maths (OTIS mock) | 71.1% | 80.0% |
| Coding | ||
| Code Migration | 6.2% | 4.6% |
| LiveCodeBench | 79.0% | 80.1% |
| ProgramBench | 0.0% | 0.0% |
| SWE-bench | 75.0% | 62.8% |
| Vibe Code Bench v1.1 | 37.2% | 0.0% |
| WeirdML | 39.0% | 52.2% |
| Agents | ||
| DTBench | 83.5% | 76.8% |
| LMCA | 37.6% | 35.0% |
| Professional work | ||
| Finance Agent (v2) | 47.4% | 30.0% |
| CorpFin v2 | 60.7% | 59.4% |
| EMB | 43.2% | 8.6% |
| LegalBench | 84.1% | 83.8% |
| Legal Research Bench | 13.9% | 3.4% |
| MedCode | 43.5% | 47.6% |
| MedScribe | 70.9% | 63.9% |
| MortgageTax | 68.7% | 68.0% |
| Tax Agent Bench | 7.2% | 3.1% |
| TaxEval v2 | 72.6% | 71.8% |
Tests at least two of these models have taken. Bold rows are the core tests.