GoogleGemma 2 27B
Best configuration ranks #499 of 610 on the BenchLeader Index at 40.9 ±5.2. Last measured 2 Sept 2026. Released 24 Jun 2024.
- Blended price
- $0.650/M
- $0.650 in · $0.650 out
- Output speed
- 54 tok/s
- First answer
- 1.34 s
- Context
- 8k
- Overall index41
- Reasoning34
- Coding42
- Maths30
- Knowledge34
- Human preference45
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 | default | Source |
|---|---|---|
| GPQA Diamond | 36.5%#237 | Epoch AI Benchmarking Hub |
| LMArena Hard Prompts | 1282#247 | LMArena |
Coding
| Benchmark | default | Source |
|---|---|---|
| LMArena Coding | 1305#253 | LMArena |
Maths
| Benchmark | default | Source |
|---|---|---|
| OTIS Mock AIME | 1.4%#245 | Epoch AI Benchmarking Hub |
| MATH Level 5 | 27.9%#65 | Epoch AI Benchmarking Hub |
Knowledge
| Benchmark | default | Source |
|---|---|---|
| LegalBench | 68.0%#121 | Vals AI |
Human preference
| Benchmark | default | Source |
|---|---|---|
| LMArena Text | 1289#238 | LMArena |
Composite
| Benchmark | default | Source |
|---|---|---|
| Epoch Capabilities Indexnot in index | 122.0#156 | Epoch AI Benchmarking Hub |
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 |
|---|---|---|---|---|---|---|
| NextBit | 54 tok/s | 1.34 s | $0.650 | $0.650 | 8k | int4 |
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. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0005 | – | 6.9 s |
| Summarise a 30-page report | 12,000 / 600 | $0.0082 | – | 12.4 s |
| Code edit | 6,000 / 1,500 | $0.0049 | – | 29.1 s |
| Agentic coding session | 60,000 / 4,000 | $0.042 | – | 1.3 min |
| Structured extraction | 2,000 / 200 | $0.0014 | – | 5.0 s |
See also
Data as of 9 Sept 2026. Compare with another model.