Fast Gemini model balancing multimodal reasoning, tool use, and cost
Specification
gemini listing
Context
1M
Max output
8K
Released
2025-08-05
Knowledge cutoff
—
Retires
2026-06-01
Open weights
no
Input
text, image, audio, video
Output
text
Price
US dollars per million tokens · gemini list price
Input
$0.10
Output
$0.40
Cache read
$0.025
Batch and priority prices
US dollars per million tokens · per listing
| Price | Listing | $/M |
|---|---|---|
| Batch input | vertex_ai-language-models | $0.075 |
| Batch output | vertex_ai-language-models | $0.30 |
Batch input is 50% below the interactive input price at the same listing.
Rate limits
as listed
Requests per minute
10,000
Tokens per minute
10,000,000
Quality
6 benchmarks · 3 sources
Composite
37th
percentile of 334 scored models
Rank
210 / 334
±0.04 sd
Evidence
6 × 3
3 or more benchmarks, from 2 or more sources
Effort range
—
one configuration only
coding16th2/3 bench
reasoning52nd1/4 bench
math51st3/7 bench
Every score
one row per source and configuration — nothing averaged away
| Benchmark | Score | Configuration | Source | Run |
|---|---|---|---|---|
| MATH level 5 | 82.2 ±0.9 | default | Epoch AI | 2025-02-06 |
| GPQA diamond | 60.6 ±3.5 | default · 2 runs | Epoch AI | 2025-02-06 |
| OTIS Mock AIME 2024-2025 | 44.4 ±7.4 | default · 2 runs | Epoch AI | 2025-03-07 |
| FrontierMath-v1 | 0.9 ±0.8 | default · 2 runs | Epoch AI | 2025-03-09 |
| Aider polyglot | 22.2 | default · aider whole | Aider | 2024-12-22 |
| SWE-Bench verified | 28.9 | default · CodeShellAgent · 2 runs | SWE-bench | — |
The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.
Available from 5 hosts
| Host | In $/M | Out $/M | Cache rd | Cache wr | Context | Output | Retires |
|---|---|---|---|---|---|---|---|
| geminivendor | $0.10 | $0.40 | $0.025 | — | 1M | 8K | 2026-06-01 |
| poe | $0.10 | $0.42 | — | — | 990K | 8K | — |
| vercel_ai_gateway · vercel-ai-gateway | $0.15 | $0.60 | — | — | 1M | 8K | 2026-06-01 |
| vertex_ai-language-models | $0.15 | $0.60 | $0.025 | — | 1M | 8K | 2026-06-01 |
| qiniu-ai | — | — | — | — | 1M | 8K | — |
Price history
append-only observations · a listing writes a row only when its price moves
| Date | Host | Field | $/M |
|---|---|---|---|
| 2026-09-02 | vertex_ai-language-models | Input | $0.10 |
| 2026-09-15 | vertex_ai-language-models | Input | $0.15 |
| 2026-09-02 | vertex_ai-language-models | Output | $0.40 |
| 2026-09-15 | vertex_ai-language-models | Output | $0.60 |
More from Google
most-hosted first