GoogleReasoning modelGemini 2.5 Flash-Lite
Best configuration ranks #363 of 610 on the BenchLeader Index at 46.7 ±4.8 (thinking reasoning effort). Last measured 2 Sept 2026. Released 17 Jun 2025.
- Blended price
- $0.175/M
- $0.100 in · $0.400 out
- Output speed
- 378 tok/s
- First answer
- 25 s
- first token 0.30 s
- Context
- 1M
- Overall index47
- Reasoning54
- Coding46
- Agents & tools38
- Knowledge42
- Instruction following51
- Human preference55
- Multimodal47
- Long context40
- Composite40
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Default” means the publisher did not say which setting was used.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| thinkingbest | 46.7 | #363 | 378 tok/s | 25 s | $0.0002 | Agents & tools 38 · Coding 46 · Composite 40 · Human preference 55 · Instruction following 51 · Knowledge 42 · Long context 40 · Multimodal 47 · Reasoning 54 |
| default | 41.6 | #481 | 290 tok/s | 0.30 s | $0.0002 | Agents & tools 44 · Coding 40 · Composite 37 · Instruction following 35 · Knowledge 36 · Long context 41 · Multimodal 38 · Reasoning 39 |
Benchmark results
One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | thinking | default | Source |
|---|---|---|---|
| LMArena Hard Prompts | 1382#164 | – | LMArena |
| GPQA Diamond (AA)not in index | 62.5%#343 | 47.4%#417 | Artificial Analysis |
| Humanity's Last Exam (AA)not in index | 6.8%#308 | 3.7%#491 | Artificial Analysis |
| Kagi LLM Benchmark | – | 40.5%#102 | Kagi LLM Benchmark |
Coding
| Benchmark | thinking | default | Source |
|---|---|---|---|
| WeirdML | 35.2%#120 | 35.2%#120 | WeirdML |
| LMArena Coding | 1383#190 | – | LMArena |
Agents & tools
| Benchmark | thinking | default | Source |
|---|---|---|---|
| Terminal-Bench Hard | 4.5%#270 | 2.3%#304 | Artificial Analysis |
| τ²-Bench Telecom (AA)not in index | 18.4%#332 | 19.0%#329 | Artificial Analysis |
| BFCL Overall | – | 36.9%#40 | Berkeley Function Calling Leaderboard |
Knowledge
| Benchmark | thinking | default | Source |
|---|---|---|---|
| AA-Omniscience | -45.6#287 | -58.8#380 | Artificial Analysis |
Instruction following
| Benchmark | thinking | default | Source |
|---|---|---|---|
| IFBench | 49.9%#172 | 31.5%#333 | Artificial Analysis |
Human preference
| Benchmark | thinking | default | Source |
|---|---|---|---|
| LMArena Text | 1375#156 | – | LMArena |
Multimodal
| Benchmark | thinking | default | Source |
|---|---|---|---|
| LMArena Vision | 1187#80 | – | LMArena |
| MMMU-Pro | 58.2%#185 | 54.0%#197 | Artificial Analysis |
Long context
| Benchmark | thinking | default | Source |
|---|---|---|---|
| Fiction.LiveBench 120k | 21.9%#33 | – | Fiction.live |
| AA-LCR | 55.7%#232 | 32.0%#315 | Artificial Analysis |
Composite
| Benchmark | thinking | default | Source |
|---|---|---|---|
| Epoch Capabilities Indexnot in index | – | 133.9#122 | Epoch AI Benchmarking Hub |
| AA Intelligence Index | 8.6#358 | 6.7#431 | Artificial Analysis |
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| Google Vertex (EU) | 154 tok/s | 0.41 s | $0.100 | $0.400 | 1.0M | – |
| Google Vertex | 90 tok/s | 1.02 s | $0.100 | $0.400 | 1.0M | – |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.010 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0002 | $0.0001 | 25.8 s |
| Summarise a 30-page report | 12,000 / 600 | $0.0014 | $0.0006 | 26.5 s |
| Code edit | 6,000 / 1,500 | $0.0012 | $0.0008 | 28.9 s |
| Agentic coding session | 60,000 / 4,000 | $0.0076 | $0.0036 | 35.5 s |
| Structured extraction | 2,000 / 200 | $0.0003 | $0.0001 | 25.5 s |
See also
Data as of 9 Sept 2026. Compare these configurations.