The following tables show the models that support
Provisioned Throughput, the throughput for each
[generative AI scale unit (GSU) and the burndown
rates](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/measure-provisioned-throughput#gsu-burndown-rate) for each
model.

## Google models

Provisioned Throughput only supports models that you call directly
from your project using the specific model ID and not a model alias. To use
Provisioned Throughput to make API calls to a model, you must use the
specific model version ID (for example, `gemini-2.0-flash-001`) and not a
[model version alias](https://docs.cloud.google.com/gemini-enterprise-agent-platform/machine-learning/model-registry/model-alias.md.txt).

While Provisioned Throughput assures capacity for your model
requests, it doesn't include or bypass quotas for other tools that you
might use, such as [Grounding](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/grounding/overview.md.txt). Depending
on the size of your workload, you might need to request additional quota for
these tools separately.

Moreover, Provisioned Throughput doesn't support
models that are called by other Gemini Enterprise Agent Platform products, such as
Agent Conversation on Gemini Enterprise Agent Platform and Agent Search on Gemini Enterprise Agent Platform. For example, if you make
API calls to Gemini 3.5 Flash while using Agent Search,
your Provisioned Throughput order for Gemini 3.5 Flash
won't guarantee the calls made by Agent Search.

Provisioned Throughput doesn't support
[batch prediction](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/capabilities/batch-prediction-gemini) calls.

The following table shows the throughput, purchase increment, and burndown rates
for Google models that support Provisioned Throughput. Your
*per-second throughput* is defined as your prompt input and generated output
across all requests per second.

To find out how many tokens your workload requires, refer to the [SDK
tokenizer](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/capabilities/list-token) or the
[countTokens API](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/capabilities/get-token-count.md.txt).

| **Model** | **Per-second throughput per GSU** | **Units** | **Minimum GSU purchase increment** | **Burndown rates** |
|---|---|---|---|---|
| [Gemini 4 Hydrogen](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/4-hydrogen.md.txt) Latest supported version: `gemini-4-hydrogen` | 685 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 1 token 1 input text cache read token = 0.05 tokens 1 input image cache read token = 0.05 tokens 1 input video cache read token = 0.05 tokens 1 input audio cache read token = 0.05 tokens 1 output response/reasoning text token = 5 tokens **Starting January 12, 2027 (by default):** 1 input text cache write token = 1.25 tokens 1 input image cache write token = 1.25 tokens 1 input video cache write token = 1.25 tokens 1 input audio cache write token = 1.25 tokens |
| [Gemini 4 Carbon](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/4-carbon.md.txt) Latest supported version: `gemini-4-carbon` | 260 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 1 token 1 input text cache read token = 0.05 tokens 1 input image cache read token = 0.05 tokens 1 input video cache read token = 0.05 tokens 1 input audio cache read token = 0.05 tokens 1 output response/reasoning text token = 5 tokens |
| [Gemini 4 Carbon for cyber](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/4-carbon.md.txt#cyber) Latest supported version: `gemini-4-carbon-cyber-permissive` | 260 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 1 token 1 input text cache read token = 0.05 tokens 1 input image cache read token = 0.05 tokens 1 input video cache read token = 0.05 tokens 1 input audio cache read token = 0.05 tokens 1 output response/reasoning text token = 5 tokens |
| [Gemini 4 Argon](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/4-argon.md.txt) Latest supported version: `gemini-4-argon` | 260 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 1 token 1 input text cache read token = 0.05 tokens 1 input image cache read token = 0.05 tokens 1 input video cache read token = 0.05 tokens 1 input audio cache read token = 0.05 tokens 1 output response/reasoning text token = 5 tokens |
| [Gemini 4 Argon Cyber Permissive](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/4-argon.md.txt#cyber) Latest supported version: `gemini-4-argon-cyber-permissive` | 260 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 1 token 1 input text cache read token = 0.05 tokens 1 input image cache read token = 0.05 tokens 1 input video cache read token = 0.05 tokens 1 input audio cache read token = 0.05 tokens 1 output response/reasoning text token = 5 tokens |
| [Gemini 3.8 Live](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-8-live.md.txt) Latest supported version: `gemini-3.8-live` | 1,350 | Tokens | 1 | 1 input text token = 1 token 1 input audio token = 4 tokens 1 input image / video token = 1.4 tokens 1 output text token = 6 tokens 1 output audio token = 16 tokens 1 output video token = 1.4 tokens |
| [Gemini 3.8 Flash Cyber](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-8-flash-cyber.md.txt) Latest supported version: `gemini-3.8-flash-cyber` | 675 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 1 token 1 input text cache read token = 0.1 tokens 1 input image cache read token = 0.1 tokens 1 input video cache read token = 0.1 tokens 1 input audio cache read token = 0.1 tokens 1 output text response token = 5 tokens |
| [Gemini 3.8 Flash](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-8-flash.md.txt) Latest supported version: `gemini-3.8-flash` | 675 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 1 token 1 input text cache read token = 0.1 tokens 1 input image cache read token = 0.1 tokens 1 input video cache read token = 0.1 tokens 1 input audio cache read token = 0.1 tokens 1 output text response token = 5 tokens |
| [Gemini Omni 1.2 Flash](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/omni-1-2-flash.md.txt) Latest supported version: `gemini-omni-1.2-flash` | 685 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 1 token 1 output text token (response \& reasoning) = 6 tokens 1 output video token = 8 tokens |
| [Gemini Omni 1.1 Flash](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/omni-1-1-flash.md.txt) Latest supported version: `gemini-omni-1.1-flash-preview` | 670 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 1 token 1 output text token (response \& reasoning) = 6 tokens 1 output video token = 12 tokens |
| [Gemini 3.7 Flash](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-7-flash.md.txt) Latest supported version: `gemini-3.7-flash` | 675 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 1 token 1 input text cache read token = 0.1 tokens 1 input image cache read token = 0.1 tokens 1 input video cache read token = 0.1 tokens 1 input audio cache read token = 0.1 tokens 1 output text response token = 5 tokens |
| [Gemini 3.6 Flash](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-6-flash.md.txt) Latest supported version: `gemini-3.6-flash` | 675 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 1 token 1 input text cache read token = 0.1 tokens 1 input image cache read token = 0.1 tokens 1 input video cache read token = 0.1 tokens 1 input audio cache read token = 0.1 tokens 1 output text response token = 5 tokens |
| [Gemini 3.5 Flash-Lite](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-5-flash-lite.md.txt) Latest supported version: `gemini-3.5-flash-lite` | 3,360 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 1 token 1 input text cache read token = 0.1 tokens 1 input image cache read token = 0.1 tokens 1 input video cache read token = 0.1 tokens 1 input audio cache read token = 0.1 tokens 1 output text response token = 9 tokens |
| [Gemini 3.1 Flash-Lite Image (Nano Banana 2 Lite)](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-1-flash-lite-image.md.txt) Latest supported version: `gemini-3.1-flash-lite-image` | 4,030 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 output response text token = 6 tokens 1 output reasoning text token = 6 tokens 1 output image token = 120 tokens |
| [Gemini 3 Pro Image](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-pro-image.md.txt) Latest supported version: `gemini-3-pro-image` | 500 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 output response text token = 6 tokens 1 output reasoning text token = 6 tokens 1 output image token = 60 tokens |
| [Gemini Nano Banana 2.1](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/nano-banana-2-1.md.txt) Latest supported version: `gemini-nano-banana-2.1` | 675 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input text cache read token = 0.1 tokens 1 input image cache read token = 0.1 tokens 1 input video cache read token = 0.1 tokens 1 output response text token = 5 tokens 1 output reasoning text token = 5 tokens 1 output image token = 20 tokens |
| [Gemini 3.1 Flash Image](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-1-flash-image.md.txt) Latest supported version: `gemini-3.1-flash-image` | 2,015 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 output response text token = 6 tokens 1 output reasoning text token = 6 tokens 1 output image token = 120 tokens |
| [Gemini 3.5 Flash](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-5-flash.md.txt) Latest supported version: `gemini-3.5-flash` | 675 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 1 token 1 input text cache read token = 0.1 tokens 1 input image cache read token = 0.1 tokens 1 input video cache read token = 0.1 tokens 1 input audio cache read token = 0.1 tokens 1 output text token = 6 tokens |
| [Gemini 3.1 Flash-Lite](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-1-flash-lite.md.txt) Latest supported version: `gemini-3.1-flash-lite` | 4030 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 2 tokens 1 input text cache read token = 0.1 tokens 1 input image cache read token = 0.1 tokens 1 input video cache read token = 0.1 tokens 1 input audio cache read token = 0.2 tokens 1 output response text token = 6 tokens 1 output reasoning text token = 6 tokens |
| [Gemini 3.1 Pro](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-1-pro.md.txt) Latest supported version: `gemini-3.1-pro-preview` ([preview](https://cloud.google.com/products#product-launch-stages)) | 500 | Tokens | 1 | **Less than or equal to 200,000 input tokens:** 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 1 token 1 input cache token = 0.1 tokens 1 output response text token = 6 tokens 1 output reasoning text token = 6 tokens <br /> **Greater than 200,000 input tokens:** 1 input text token = 2 tokens 1 input image token = 2 tokens 1 input video token = 2 tokens 1 input audio token = 2 tokens 1 input cache token = 0.2 tokens 1 output response text token = 9 tokens 1 output reasoning text token = 9 tokens |
| [Gemini 3 Flash](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-flash.md.txt) Latest supported version: `gemini-3-flash-preview` ([preview](https://cloud.google.com/products#product-launch-stages)) | 2015 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 2 tokens 1 input text, image, video cache read token = 0.1 tokens 1 input audio cache read token = 0.2 tokens 1 output response text token = 6 tokens 1 output reasoning text token = 6 tokens |
| [Gemini 2.5 Pro](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/2-5-pro.md.txt) Latest supported version: `gemini-2.5-pro` | 650 | Tokens | 1 | **Less than or equal to 200,000 input tokens:** 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 1 token 1 output response text token = 8 tokens 1 output reasoning text token = 8 tokens <br /> **Greater than 200,000 input tokens:** 1 input text token = 2 tokens 1 input image token = 2 tokens 1 input video token = 2 tokens 1 input audio token = 2 tokens 1 output response text token = 12 tokens 1 output reasoning text token = 12 tokens |
| [Gemini 2.5 Flash Image](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/2-5-flash-image.md.txt) Latest supported version: `gemini-2.5-flash-image` | 2,690 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 output text token = 9 tokens 1 output image token = 100 tokens |
| [Gemini 2.5 Flash](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/2-5-flash.md.txt) Latest supported version: `gemini-2.5-flash` | 2690 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 4 tokens 1 output response text token = 9 tokens 1 output reasoning text token = 9 tokens |
| [Gemini 2.5 Flash-Lite](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/2-5-flash-lite.md.txt) Latest supported version ([GA](https://cloud.google.com/products#product-launch-stages)): `gemini-2.5-flash-lite` Latest supported version ([preview](https://cloud.google.com/products#product-launch-stages)): `gemini-2.5-flash-lite-preview-09-2025` | 8,070 | Tokens | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 input video token = 1 token 1 input audio token = 3 tokens 1 output response text token = 4 tokens 1 output reasoning text token = 4 tokens |
| [Gemini 2.5 Flash with Gemini Live API native audio](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/2-5-flash-live-api.md.txt) Latest supported version: `gemini-live-2.5-flash-native-audio` | 1,620 | Tokens | 1 | 1 input text token = 1 token 1 input audio token = 6 tokens 1 input video token = 6 tokens 1 input image token = 6 tokens 1 input session memory token = 1 token 1 output text token = 4 tokens 1 output audio token = 24 tokens |
| [Virtual Try-On](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/vto/virtual-try-on-001.md.txt) Latest supported version: `virtual-try-on-001` | 0.02 | Images | 1 | 1 images per second = 0.02 tokens |
| [Veo 3.1 Lite Generate](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/veo/3-1-generate.md.txt#3.1-lite-generate-001-preview) Latest supported version: `veo-3.1-lite-generate-001` | 0.0350 | Video seconds (720p) | 1 | 1 (720p) output video second = 1 output video second |
| [Veo 3.1 Lite Generate](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/veo/3-1-generate.md.txt#3.1-lite-generate-001-preview) Latest supported version: `veo-3.1-lite-generate-001` | 0.0350 | Video+audio seconds (720p) | 1 | 1 (720p) output video+audio second = 1.75 output video seconds |
| [Veo 3.1 Lite Generate](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/veo/3-1-generate.md.txt#3.1-lite-generate-001-preview) Latest supported version: `veo-3.1-lite-generate-001` | 0.0350 | Video seconds (1080p) | 1 | 1 (1080p) output video second = 1.75 (720p) output video second |
| [Veo 3.1 Lite Generate](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/veo/3-1-generate.md.txt#3.1-lite-generate-001-preview) Latest supported version: `veo-3.1-lite-generate-001` | 0.0350 | Video+audio seconds (1080p) | 1 | 1 (1080p) output video+audio second = 2.33 (720p) output video seconds |
| [Veo 3.1](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/veo/3-1-generate.md.txt#3.1-generate-001) Latest supported version: `veo-3.1-generate-001` | 0.0040 | Video seconds | 1 | 1 output video second = 1 output video second |
| [Veo 3.1](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/veo/3-1-generate.md.txt#3.1-generate-001) Latest supported version: `veo-3.1-generate-001` | 0.0040 | Video+audio seconds | 1 | 1 output video+audio second = 2 output video seconds |
| [Veo 3.1 Fast](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/veo/3-1-generate.md.txt#3.1-fast-generate-001) Latest supported version: `veo-3.1-fast-generate-001` | 0.01 | Video seconds (720p) | 1 | 1 (720p) output video sec = 1 output video second |
| [Veo 3.1 Fast](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/veo/3-1-generate.md.txt#3.1-fast-generate-001) Latest supported version: `veo-3.1-fast-generate-001` | 0.01 | Video + Audio seconds (720p) | 1 | 1 (720p) output video + audio second = 1.30 output video seconds |
| [Veo 3.1 Fast](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/veo/3-1-generate.md.txt#3.1-fast-generate-001) Latest supported version: `veo-3.1-fast-generate-001` | 0.01 | Video seconds (1080p) | 1 | 1 (1080p) output video second = 1.30 (720p) output video seconds |
| [Veo 3.1 Fast](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/veo/3-1-generate.md.txt#3.1-fast-generate-001) Latest supported version: `veo-3.1-fast-generate-001` | 0.01 | Video + Audio seconds (1080p) | 1 | 1 (1080p) output video + audio second = 1.60 (720p) output video seconds |
| [Veo 3.1 Fast](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/veo/3-1-generate.md.txt#3.1-fast-generate-001) Latest supported version: `veo-3.1-fast-generate-001` | 0.01 | Video seconds (4k) | 1 | 1 (4k) output video second = 3.40 (720p) output video seconds |
| [Veo 3.1 Fast](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/veo/3-1-generate.md.txt#3.1-fast-generate-001) Latest supported version: `veo-3.1-fast-generate-001` | 0.01 | Video + Audio seconds (4k) | 1 | 1 (4k) output video + audio second = 4 (720p) output video seconds |

For information about a model's capabilities and input or output limits,
see the documentation for the model.

> [!NOTE]
> **Note:** Starting on January 12, 2027, [Gemini 4 Hydrogen](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/4-hydrogen.md.txt) and later models support cache control features with TTL configuration. For these models, cache writes burn down at 1.25 tokens per token by default, and longer retention burns down at a higher rate.

**Request access:** The model `gemini-live-2.5-flash` is in private GA. For information about access to this release, see the [access request page](https://docs.google.com/forms/d/e/1FAIpQLScxBeD4UJ8GbUfX4SXjj5a1XJ1K7Urwvb0iSGdGccNcFRBrpQ/viewform).

You can upgrade to new models as they are made available. For information about
model availability and discontinuation dates, see [Google
models](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/model-versions#stable-versions-available).

For more information about supported locations, see
[Available locations](https://docs.cloud.google.com/gemini-enterprise-agent-platform/resources/locations.md.txt#available-regions).

## Partner models

The following table shows the throughput, purchase increment, and burndown rates
for [partner models](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/partner-models/use-partner-models.md.txt) that
support Provisioned Throughput. [Claude models](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/supported-models.md.txt#partner-models) are
measured in tokens per second, which is defined as a total of input and output
tokens across all requests per second.

| **Model** | **Throughput per GSU (tokens/sec)** | **Minimum GSU purchase** | **GSU purchase increment** | **Burndown rates** |
|---|---|---|---|---|
| Anthropic's Anthropic's Claude Haiku 5.5 on Google Cloud | 10,417 | 1 | 1 | **100,000 input tokens or fewer:** 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache write 1h token = 2 tokens 1 cache hit token = 0.1 token **More than 100,000 input tokens:** 1 input token = 5 tokens 1 output token = 25 tokens 1 cache write 5m token = 6.25 tokens 1 cache write 1h token = 10 tokens 1 cache hit token = 0.5 token |
| Anthropic's Anthropic's Claude Sonnet 5.5 on Google Cloud | 525 | 1 | 1 | 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache write 1h token = 2 tokens 1 cache hit token = 0.05 token |
| Anthropic's Anthropic's Claude Opus 5.5 on Google Cloud | 261 | 1 | 1 | 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache write 1h token = 2 tokens 1 cache hit token = 0.05 token |
| Anthropic's Claude Fable 5.2 on Google Cloud [Fast mode](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/partner-models/claude/fast-mode.md.txt) (`"speed": "fast"`) isn't covered by Provisioned Throughput and is always billed as [pay-as-you-go](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/standard-paygo.md.txt). | 131 | 1 | 1 | 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache write 1h token = 2 tokens 1 cache hit token = 0.025 token |
| Anthropic's Anthropic's Claude Fable 5.1 on Google Cloud | 105 | 1 | 1 | 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache write 1h token = 2 tokens 1 cache hit token = 0.025 token |
| Anthropic's Anthropic's Claude Opus 5 on Google Cloud | 210 | 1 | 1 | 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache write 1h token = 2 tokens 1 cache hit token = 0.1 token |
| Anthropic's Anthropic's Claude Sonnet 5 on Google Cloud | 350 | 25 | 1 | 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache write 1h token = 2 tokens 1 cache hit token = 0.1 token |
| Anthropic's Anthropic's Claude Fable 5 on Google Cloud | 105 | 1 | 1 | 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache write 1h token = 2 tokens 1 cache hit token = 0.1 token |
| Anthropic's Anthropic's Claude Opus 4.8 on Google Cloud | 210 | 35 | 1 | 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache write 1h token = 2 tokens 1 cache hit token = 0.1 token |
| Anthropic's Anthropic's Claude Opus 4.7 on Google Cloud | 210 | 35 | 1 | 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache write 1h token = 2 tokens 1 cache hit token = 0.1 token |
| Anthropic's Anthropic's Claude Sonnet 4.6 on Google Cloud | 350 | 25 | 1 | 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache write 1h token = 2 tokens 1 cache hit token = 0.1 token |
| Anthropic's Anthropic's Claude Opus 4.6 on Google Cloud | 210 | 35 | 1 | 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache write 1h token = 2 tokens 1 cache hit token = 0.1 token |
| Anthropic's Anthropic's Claude Opus 4.5 on Google Cloud | 210 | 35 | 1 | 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache write 1h token = 2 tokens 1 cache hit token = 0.1 token |
| Anthropic's Claude Sonnet 4.5 on Google Cloud (deprecated) | 350 | 25 | 1 | **Less than 200,000 input tokens:** 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache write 1h token = 2 tokens 1 cache hit token = 0.1 token **Greater than or equal to 200,000 input tokens:** 1 input token = 2 token 1 output token = 7.5 tokens 1 cache write 5m token = 2.5 tokens 1 cache write 1h token = 4 tokens 1 cache hit token = 0.2 token |
| Anthropic's Anthropic's Claude Haiku 4.5 on Google Cloud | 1,050 | 8 | 1 | **Less than 200,000 input tokens:** 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache write 1h token = 2 tokens 1 cache hit token = 0.1 token |
| Anthropic's Claude 3.7 Sonnet on Google Cloud (deprecated) | 350 | 25 | 1 | 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache hit token = 0.1 token |
| Anthropic's Claude 3.5 Sonnet v2 on Google Cloud (deprecated) | 350 | 25 | 1 | 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache hit token = 0.1 token |
| Anthropic's Claude 3.5 Haiku on Google Cloud (deprecated) | 2,000 | 10 | 1 | 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache write 1h token = 2 tokens 1 cache hit token = 0.1 token |
| Anthropic's Claude 3 Opus on Google Cloud | 70 | 35 | 1 | 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache hit token = 0.1 token |
| Anthropic's Claude 3 Haiku on Google Cloud (deprecated) | 4,200 | 5 | 1 | 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache write 1h token = 2 tokens 1 cache hit token = 0.1 token |
| Anthropic's Claude 3.5 Sonnet on Google Cloud (deprecated) | 350 | 25 | 1 | 1 input token = 1 token 1 output token = 5 tokens 1 cache write 5m token = 1.25 tokens 1 cache hit token = 0.1 token |

For information about supported locations, see
[Anthropic Claude region availability](https://docs.cloud.google.com/gemini-enterprise-agent-platform/resources/locations.md.txt#genai-partner-models).
To order Provisioned Throughput for Anthropic models, contact your
[Google Cloud account representative](https://cloud.google.com/contact).

## Open models

> [!WARNING]
>
> **Preview**
>
>
> This feature is
>
> subject to the "Pre-GA Offerings Terms" in the General Service Terms section of the
> [Service Specific
> Terms](https://docs.cloud.google.com/terms/service-terms#1).
>
> Pre-GA features are available "as is" and might have limited support.
>
> For more information, see the
> [launch stage descriptions](https://cloud.google.com/products/#product-launch-stages).

The following table shows the throughput, purchase increment, and burndown rates
for [open models](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/maas/use-open-models.md.txt) that
support Provisioned Throughput.

| **Model** | **Throughput per GSU (tokens/sec)** | **Minimum GSU purchase** | **GSU purchase increment** | **Burndown rates** |
|---|---|---|---|---|
| [Gemma 4 26B A4B IT](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/maas/google/gemma-4-26b-a4b-it.md.txt) Latest supported version: `gemma-4-26b-a4b-it-maas` Provisioned Throughput support for this model is billed under the same SKUs as Google models, but is subject to open model [capabilities](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/supported-models.md.txt#capabilities). | 6,725 | 1 | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 output text token = 4 tokens |
| [Llama 4 Maverick 17B-128E](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/partner-models/llama/llama4-maverick.md.txt) Latest supported version: `llama-4-maverick-17b-128e-instruct-maas` | 2,800 | 1 | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 output text token = 4 tokens |
| [Llama 4 Scout 17B-16E](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/partner-models/llama/llama4-scout.md.txt) Latest supported version: `llama-4-scout-17b-16e-instruct-maas` | 4,035 | 1 | 1 | 1 input text token = 1 token 1 input image token = 1 token 1 output text token = 3 tokens |
| [OpenAI gpt-oss 120B](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/maas/openai/gpt-oss-120b.md.txt) Latest supported version: `gpt-oss-120b-maas` | 11,205 | 1 | 1 | 1 input text token = 1 token 1 output text token = 4 tokens |

## Available capabilities for Google and open models

The following table lists the capabilities
that are available with Provisioned Throughput for Google models and open models:

| Capability | Google models | Open models (preview) |
|---|---|---|
| Order through Google Cloud console | Yes | Yes |
| Supports global endpoints | See [Global endpoint model support](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/supported-models.md.txt#global-endpoint). | See [Global endpoint model support](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/supported-models.md.txt#global-endpoint). |
| Supports [supervised fine-tuned models](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/supported-models.md.txt#supervised-fine-tuned-model-support) | Yes | No |
| Supports [API key usage](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/use-provisioned-throughput.md.txt#api-key) | Yes | No |
| Integrated with [implicit context caching](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/measure-provisioned-throughput.md.txt#context_caching) | Yes | Not applicable |
| Integrated with explicit context caching | Yes | Not applicable |
| ML processing | Available in specific regions. For details, see [Single Zone Provisioned Throughput](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/szpt.md.txt). | Not applicable |
| Available order terms | 1 week, 1 month, 3 month, and 1 year | 1 week, 1 month, 3 month, and 1 year |
| [Change order](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/purchase-provisioned-throughput.md.txt#change-order) from the console | Yes | No |
| [Order statuses](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/purchase-provisioned-throughput.md.txt#check-order-status) | Pending review, Approved, Declined, Active, Cancelled, Terminated, Expired | Pending review, Approved, Declined, Active, Expired |
| Overages [spillover](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/use-provisioned-throughput.md.txt#default) to pay-as-you-go by default | Yes | Yes |
| [API header control](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/use-provisioned-throughput.md.txt#use-rest-api): use "dedicated" to only use provisioned throughput or "shared" to only use pay-as-you-go | Yes | Yes |
| [Monitoring](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/use-provisioned-throughput.md.txt#monitor_provisioned_throughput): [metrics](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/use-provisioned-throughput.md.txt#metrics), [dashboards](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/use-provisioned-throughput.md.txt#dashboards), and [alerting](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/use-provisioned-throughput.md.txt#alerts) | Yes | Yes |

## Global endpoint model support

Provisioned Throughput supports the
[global endpoint](https://docs.cloud.google.com/gemini-enterprise-agent-platform/resources/locations.md.txt#global-endpoint) for Google models and open models.

Traffic that exceeds the Provisioned Throughput quota uses the
global endpoint, by default.

To assign Provisioned Throughput to the global endpoint of a model,
select `global` as the region when you [place a Provisioned Throughput order](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/purchase-provisioned-throughput.md.txt#place-an-order).

### Google models with global endpoint support

The following table lists the Google models for which Provisioned Throughput supports the global endpoint:

| Model | Latest supported model version |
|---|---|
| [Gemini 4 Hydrogen](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/4-hydrogen.md.txt) | `gemini-4-hydrogen` |
| [Gemini 4 Carbon](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/4-carbon.md.txt) | `gemini-4-carbon` |
| [Gemini 4 Carbon for cyber](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/4-carbon.md.txt#cyber) | `gemini-4-carbon-cyber-permissive` |
| [Gemini 4 Argon](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/4-argon.md.txt) | `gemini-4-argon` |
| [Gemini 4 Argon Cyber Permissive](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/4-argon.md.txt#cyber) | `gemini-4-argon-cyber-permissive` |
| [Gemini 3.8 Flash Cyber](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-8-flash-cyber.md.txt) | `gemini-3.8-flash-cyber` |
| [Gemini 3.8 Flash](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-8-flash.md.txt) | `gemini-3.8-flash` |
| [Gemini Omni 1.2 Flash](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/omni-1-2-flash.md.txt) | `gemini-omni-1.2-flash` |
| [Gemini Omni 1.1 Flash](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/omni-1-1-flash.md.txt) preview | `gemini-omni-1.1-flash-preview` |
| [Gemini 3.7 Flash](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-7-flash.md.txt) | `gemini-3.7-flash` |
| [Gemini 3.6 Flash](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-6-flash.md.txt) | `gemini-3.6-flash` |
| [Gemini 3.5 Flash-Lite](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-5-flash-lite.md.txt) | `gemini-3.5-flash-lite` |
| [Gemini 3.1 Flash-Lite Image (Nano Banana 2 Lite)](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-1-flash-lite-image.md.txt) | `gemini-3.1-flash-lite-image` |
| [Gemini 3 Pro Image](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-pro-image.md.txt) | `gemini-3-pro-image` |
| [Gemini Nano Banana 2.1](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/nano-banana-2-1.md.txt) | `gemini-nano-banana-2.1` |
| [Gemini 3.1 Flash Image](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-1-flash-image.md.txt) | `gemini-3.1-flash-image` |
| [Gemini 3.5 Flash](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-5-flash.md.txt) | `gemini-3.5-flash` |
| [Gemini 3.1 Flash-Lite](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-1-flash-lite.md.txt) | `gemini-3.1-flash-lite` |
| [Gemini 3.1 Pro](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-1-pro.md.txt) preview | `gemini-3.1-pro-preview` |
| [Gemini 3 Flash](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-flash.md.txt) preview | `gemini-3-flash-preview` |
| [Gemini 2.5 Pro](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/2-5-pro.md.txt) | `gemini-2.5-pro` |
| [Gemini 2.5 Flash Image](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/2-5-flash-image.md.txt) | `gemini-2.5-flash-image` |
| [Gemini 2.5 Flash](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/2-5-flash.md.txt) | `gemini-2.5-flash` |
| [Gemini 2.5 Flash-Lite](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/2-5-flash-lite.md.txt) | `gemini-2.5-flash-lite` |

### Open models with global endpoint support

> [!WARNING]
>
> **Preview**
>
>
> This feature is
>
> subject to the "Pre-GA Offerings Terms" in the General Service Terms section of the
> [Service Specific
> Terms](https://docs.cloud.google.com/terms/service-terms#1).
>
> Pre-GA features are available "as is" and might have limited support.
>
> For more information, see the
> [launch stage descriptions](https://cloud.google.com/products/#product-launch-stages).

The following table lists the open models for which Provisioned Throughput supports the global endpoint:

| **Model** | **Latest supported model version** |
|---|---|
| OpenAI gpt-oss 120B | `gpt-oss-120b-maas` |

## Supervised fine-tuned model support

Provisioned Throughput that you purchase for a Google base model also
covers supervised fine-tuned versions of that model, which share the base
model's quota. This applies to Google [models that support supervised
fine-tuning](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/tuning/supervised-tuning/use#expandable-1):

- Provisioned Throughput can be applied to both base models and supervised fine-tuned versions of those base models.
- Supervised fine-tuned model endpoints and their corresponding base model count toward the same Provisioned Throughput quota. For example, Provisioned Throughput purchased for `gemini-3.1-flash-lite` for a specific project prioritizes requests that are made from supervised fine-tuned versions of `gemini-3.1-flash-lite` created within that project. [Use the
  appropriate header](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/use-provisioned-throughput#use-rest-api) to control traffic behavior.

Burndown rates for fine-tuned model inference might be higher than for base
model inference. If the model has a
[fine-tuned inference premium](https://cloud.google.com/gemini-enterprise-agent-platform/generative-ai/pricing#model-tuning), the
burndown rate includes a premium that's directly proportional to it. For
example, if fine-tuned inference costs 50% more than base model inference, then
the burndown rate for fine-tuned inference is 50% higher: 1 input text token
burns down 1 token for the base model and 1.5 tokens for the fine-tuned model.

## What's next

- [Calculate Provisioned Throughput requirements](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/provisioned-throughput/measure-provisioned-throughput.md.txt).