Gemini 2.5 Flash-Lite Preview 09-2025 Thinking
| Released | 2025-09-25 |
|---|---|
| Price per 1M | $0.10 in · $0.40 out |
| Overall score | 46.5 · rank 313 |
| Context window | 1,048,576 tokens |
| Provider | |
| Weights | Proprietary |
| Configuration | Thinking |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 46.5 | 313 | 4 |
| Overall: Coding | 48.0 | 287 | 2 |
| Overall: Reasoning | 44.6 | 336 | 2 |
| Overall: Agentic work | 47.4 | 303 | 2 |
| Overall: Frontend | 48.4 | 240 | 2 |
| Coding | 48.5 | 260 | 1 |
| Reasoning | 45.7 | 316 | 1 |
| Knowledge | 42.1 | 176 | 1 |
| Multimodal | 48.3 | 121 | 1 |
| Value vs price | +1.2 | 210 |
Evidence
Coding 48.5 ±5.3
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LiveCodeBench (Vals run) | 71.4% | Vals AI | 2026-09-01 |
Reasoning 45.7 ±4.8
| Benchmark | Result | Source | Date |
|---|---|---|---|
| GPQA Diamond (Vals run) | 70.2% | Vals AI | 2026-09-01 |
Knowledge 42.1 ±4.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MMLU-Pro (Vals run) | 79.1% | Vals AI | 2026-09-01 |
Multimodal 48.3 ±2.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MMMU-Pro (Vals run) | 75.4% | Vals AI | 2026-09-01 |
Names matched to this model
gemini-2-5-flash-lite-preview-09-2025-thinking
Compare Gemini 2.5 Flash-Lite Preview 09-2025 Thinking with other models