Fast Gemini model balancing multimodal reasoning, tool use, and costGemini 3.5 Flash by google is offered by 41 providers on this page. Its official list price is $1.50 per 1M input tokens and $9.00 per 1M output tokens. The lowest paid channel is UnoRouter at $0.186 / $1.11 per 1M, about 8.1× below the list price. That channel is a third-party gateway, so confirm its availability and rate limits before depending on it.
Median output speed is 201 tokens per second, with 15.74s to the first token.
The context window is 1,048,576 tokens at the reference host, but hosts report different limits, from 200,000 to 1,048,576, so the usable window depends on the provider you pick. It supports reasoning, tool use, and structured output. Accepted input modalities are Text, Image, Audio, Video, and PDF. Providers report the capability flags inconsistently, so verify a specific feature against the host you plan to use. Its training knowledge cuts off at 2025-01.
Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.
| Provider | Tier | Input | Output | Cache read | Cache write | Context | Output limit | Status |
|---|---|---|---|---|---|---|---|---|
| UnoRouter | Gateway | $0.19 | $1.11 | — | — | 1,048,576 | 65,536 | |
| AIHubMix | Gateway | $0.75 | $3.75 | $0.075 | — | 1,000,000 ⚠ | 64,000 | deprecated |
| VertexOfficial | Cloud | $1.50 | $9.00 | $0.15 | — | 1,048,576 | 65,536 | |
| GoogleOfficial | First-party | $1.50 | $9.00 | $0.15 | — | 1,048,576 | 65,536 | |
| CrossModel | Gateway | $1.50 | $9.00 | $0.15 | $1.50 | 1,048,576 | 65,536 | |
| LLM Gateway | Gateway | $1.50 | $9.00 | $0.15 | $0.083 | 1,048,576 | 65,536 | |
| CommandCode | Gateway | $1.50 | $9.00 | $0.15 | — | 1,000,000 ⚠ | 65,536 | |
| 302.AI | Gateway | $1.50 | $9.00 | — | — | 1,048,576 | 65,536 | |
| 302.AI gemini-3.5-flash-thinking | Gateway | $1.50 | $9.00 | — | — | 1,048,576 | 65,536 | |
| Requesty | Gateway | $1.50 | $9.00 | $0.15 | $1.58 | 1,048,576 | 65,535 |
Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.
Interleaved thinking (reasoning between tool calls) is declared by 2 of 41 providers.
5 / 41 providers expose no reasoning control (reasoning_options: []).
| Provider | Audio in | Audio out |
|---|---|---|
| AIHubMix | $0.751.0× | — |
| Vertex | $1.501.0× | — |
| $1.501.0× | — | |
| LLM Gateway | $3.002.0× | — |
| Databricks | $1.501.0× | — |
| NEAR AI Cloud | $1.501.0× | — |
| Merge Gateway | $1.501.0× | — |
| Neon | $1.501.0× | — |
| OpenRouter | $3.002.0× | — |
| SAP AI Core | $1.501.0× | — |
| GitHub Copilot | $1.501.0× | — |
| Eden AI | $3.002.0× | — |
| Eden AI | $3.002.0× | — |
| Ofox | $1.501.0× | — |
| OrcaRouter | $3.002.0× | — |
| OpenCode | $1.501.0× | — |
| Eden AI | $3.302.0× | — |
| Eden AI | $3.302.0× | — |
Audio tokens are priced on a separate scale from text tokens, so these are not comparable with the text prices above. The multiplier compares each price with the same provider's text price for the same direction.
| Mode | Provider | Input | Output | Cache read | Cache write |
|---|---|---|---|---|---|
| fast | Databricks | $2.70 | $16.20 | $0.27 | — |
| Name | Conditions | Score | Metric | Source |
|---|---|---|---|---|
| Terminal-Bench | harness: Terminus-2 · v2.1 | 76.2 | success rate | Source ↗ |
| SWE-Bench Pro | variant: single attempt · dataset: public | 55.1 | resolve rate | Source ↗ |
| MCP Atlas | — | 83.6 | success rate | Source ↗ |
| Toolathlon | — | 56.5 | success rate | Source ↗ |
| OSWorld-Verified | — | 78.4 | success rate | Source ↗ |
| MMMU Pro | variant: no tools | 83.6 | accuracy | Source ↗ |
| CharXiv Reasoning | variant: no tools | 84.2 | accuracy | Source ↗ |
| Humanity's Last Exam | dataset: full set, text + MM | 40.2 | accuracy | Source ↗ |
| ARC-AGI-2 | — | 72.1 | accuracy | Source ↗ |
| GDPval-AA | — | 1656 | Elo | Source ↗ |
The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.