How much does the Gemma 4 26B A4B API cost?

On FastMetal, Gemma 4 26B A4B is billed per token: $0.0739 per 1M input tokens and $0.3587 per 1M output tokens on a US-dollar account, before tax. Usage is drawn from a prepaid balance; there is no subscription or monthly fee.

Can I call Gemma 4 26B A4B with the OpenAI SDK?

Yes. Point base_url at https://api.fastmetal.ai/v1 and pass "gemma-4-26b-a4b-it" as the model; existing OpenAI-style code works unchanged, including streaming, tool calls and structured output.

What is the context window of Gemma 4 26B A4B ?

262,144 tokens, shared between the prompt and the response.

What do I need to try Gemma 4 26B A4B ?

Create an account and add credit; the model is then available both in the browser chat and over the API. There is no contract or minimum spend.

Back to Models

Google

gemma-4-26b-a4b-it

Gemma 4 26B A4B

Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate per token during inference — delivering near-31B quality at...

4/3/2026

262,144 tokens

Input: $0.0739/M

Output: $0.3587/M

Try in Chat Get API Key

Specifications

Modalities

Input

image

text

video

Output

text

Supported Parameters

frequency_penalty

include_reasoning

logit_bias

logprobs

max_tokens

min_p

presence_penalty

reasoning

repetition_penalty

response_format

seed

stop

structured_outputs

temperature

tool_choice

tools

top_k

top_logprobs

top_p

Max Output Tokens

262,144

Reasoning Configuration

Default: Thinking off
Turning thinking off: Possible (reasoning.enabled: false)

How to control thinking, and what it costs

Data policy

Prompt retention: Unknown — we could not confirm
Training: Not used for training

"Unknown" does not mean "safe". It means we could not confirm it.

Whether a provider trains on prompts is a declared value from our terms with them. Retention is determined from the upstream listing for every host this model can reach. Neither is guessed.

Measured performance

Time to first response (p50)371ms

Throughput (p50)68 tokens/s

Measured through FastMetal's own gateway over the last 30 days (660 first-response samples).

Throughput is sampled once a day: the median of 26 observations over the last 30 days.

Time to first response is measured to the first streamed chunk; throughput counts all output tokens, including reasoning tokens. These are observations, not a performance guarantee.

Code Examples

curl https://api.fastmetal.ai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "gemma-4-26b-a4b-it",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

How Gemma 4 26B A4B actually answers

Real responses to our standard prompts, recorded on FastMetal.

Count the number of 'r's in 'strawberry'

Count the number of 'r's in 'strawberry'. Explain your reasoning step by step.

To count the number of 'r's in the word "strawberry," we can follow these steps: 1. **Identify the target letter:** We are looking specifically for the lowercase letter **'r'**. 2. **Break the word into individual letters:** s — t — **r** — a — w — b — e — **r** — **r** — y 3. **Scan the sequence for the target letter:** * 1st letter: s (No) * 2nd letter: t (No) * **3rd letter: r (Yes - 1)** * 4th letter: a (No) * 5th letter: w (No) * 6th letter: b (No) * 7th letter: e (No) * **8th letter: r (Yes - 2)** * **9th letter: r (Yes - 3)** * 10th letter: y (No) **Total Count:** There are **3** 'r's in "strawberry".

1.8s response249 output tokens

Debug This Error

I'm getting the following error in my Node.js application: TypeError: Cannot read properties of undefined (reading 'map') at UserList (/app/components/UserList.js:12:25) at renderWithHooks (/app/node_modules/rea…

### The Cause The error `TypeError: Cannot read properties of undefined (reading 'map')` means that when your `UserList` component is being rendered, the `users` prop is **`undefined`**. In JavaScript, you cannot call a method (like `.map()`) on something that doesn't exist. This usually happens for one of three reasons: 1. **Asynchronous Data Fetching:** You are fetching the user data from an API. When the component first renders, the data hasn't arrived yet, so `users` is undefined. 2. **Missing Prop:** The parent component calling `<UserList />` forgot to pass the `users` prop entirely. 3. **Incorrect Prop Name:** The parent is passing the data under a different name (e.g., `<UserLis…

21.0s response778 output tokens

Code Review

Please review the following Python function and suggest improvements for readability, performance, and best practices: def get_data(url, retries=3): import requests import time for i in range(retries):…

The provided function works, but it has several issues regarding **error handling**, **efficiency**, and **coding standards**. Here is a review of the issues and a refactored version of the code. ### Key Areas for Improvement 1. **Imports inside the function:** In Python, imports should be at the top of the file. Importing inside a function makes it slower if the function is called repeatedly and violates PEP 8 standards. 2. **Bare `except` clause:** The `except:` block catches *everything*, including `KeyboardInterrupt` (Ctrl+C) and `SystemExit`. This makes it very difficult to stop a running script. You should catch specific exceptions like `requests.exceptions.RequestException`. 3.…

8.5s response1088 output tokens

Compare these answers side by side with other models →

Frequently asked questions

How much does the Gemma 4 26B A4B API cost?: On FastMetal, Gemma 4 26B A4B is billed per token: $0.0739 per 1M input tokens and $0.3587 per 1M output tokens on a US-dollar account, before tax. Usage is drawn from a prepaid balance; there is no subscription or monthly fee.
Can I call Gemma 4 26B A4B with the OpenAI SDK?: Yes. Point base_url at https://api.fastmetal.ai/v1 and pass "gemma-4-26b-a4b-it" as the model; existing OpenAI-style code works unchanged, including streaming, tool calls and structured output.
What is the context window of Gemma 4 26B A4B ?: 262,144 tokens, shared between the prompt and the response.
What do I need to try Gemma 4 26B A4B ?: Create an account and add credit; the model is then available both in the browser chat and over the API. There is no contract or minimum spend.