Scheduled for retirement on Mar 30, 2027. Still available at the prices below until then.
Context
131K
Max output
131K
Serving providers
3
Cheapest input
$0.04 /1M
On-demand listed rates, OpenRouter routed quotes and Vercel reseller quotes. Enterprise commitments (committed-use discounts, provisioned throughput, negotiated rates) can be materially lower.
Benchmarks
independent evaluations · source: Artificial AnalysisHeadline indices
Individual benchmarks · click a row for effort variants
| Benchmark | Best score | Effort | Cost / task | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CritPt | 0% | default | — | |||||||||||||
| EvalComputeProxyFrozen Aug 18, 2026 | 0.8 | default | — | |||||||||||||
| GPQA Diamond | 29% | default | — | |||||||||||||
| ||||||||||||||||
| Humanity's Last Exam | 5% | default | — | |||||||||||||
| IFBench | 28% | default | — | |||||||||||||
| LiveCodeBench | 11% | default | — | |||||||||||||
| ||||||||||||||||
| Long-Context Reasoning | 7% | default | — | |||||||||||||
| MMLU-ProFrozen Aug 18, 2026 | 42% | default | — | |||||||||||||
| ||||||||||||||||
| MMMU-Pro | 30% | default | — | |||||||||||||
| Omniscience | -82.9 | default | — | |||||||||||||
| Omniscience Accuracy | 0.1 | default | — | |||||||||||||
| Omniscience Non Hallucination | 0.0 | default | — | |||||||||||||
| SciCodeArtificial Analysis did not report this score in its latest run on Oct 14, 2026 | 7% | — | — | |||||||||||||
| TerminalBench Hard | 1% | default | — | |||||||||||||
| TerminalBench v2.1 | 0% | default | — | |||||||||||||
| ||||||||||||||||
| τ-bench Banking | 0% | default | — | |||||||||||||
| τ²-bench | 5% | default | — | |||||||||||||
Artificial Analysis last published scores for this benchmark on Sep 9, 2026. Scores here are frozen at that date, and models released since are not on it.
Artificial Analysis last published scores for this benchmark on Aug 18, 2026. Scores here are frozen at that date, and models released since are not on it.
Pricing by serving provider
standard tier · per 1M tokens · click a provider to expandInput pricing runs from $0.04 to $0.05 per 1M tokens across 3 providers, so the dearest route costs 25% more than the cheapest for the same model.
| Serving provider | Input /1M | Output /1M | Endpoints | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Amazon BedrockCheapest | $0.04 | $0.08 | 1 | ||||||||||
Standardgoogle.gemma-3-4b-it
| |||||||||||||
| DeepInfra | $0.05 | $0.10 | 1 | ||||||||||
Standarddeepinfra/google/gemma-3-4b-it
| |||||||||||||
| OpenRouter | $0.05 | $0.10 | 1 | ||||||||||
Standardgoogle/gemma-3-4b-it
| |||||||||||||
Price history
listed seller prices, input and output per 1M tokens Input Output
Cost calculator
estimate your monthly spend on this model$14
estimated / month
Model IDs
copy the exact identifier for your platformgoogle.gemma-3-4b-itgoogle/gemma-3-4b-itdeepinfra/google/gemma-3-4b-it
Frequently asked questions
Gemma 3 4B pricing, context and availabilityIs Gemma 3 4B still available?
Yes. Gemma 3 4B is scheduled for retirement on 2027-03-30.
How much does Gemma 3 4B cost?
Gemma 3 4B costs $0.04 per 1M input tokens from Amazon Bedrock, and $0.08 per 1M output tokens from Amazon Bedrock. Across 3 serving providers, input prices range from $0.04 to $0.05 per 1M tokens.
What is the context window of Gemma 3 4B?
Gemma 3 4B accepts up to 131K tokens of context and can return up to 131K output tokens.
Which providers serve Gemma 3 4B?
Gemma 3 4B is available from 3 serving providers, each with its own pricing and model ID. Amazon Bedrock is currently the cheapest.
What can Gemma 3 4B do?
Gemma 3 4B supports image input (vision) and tool use.
Other Google models
compare pricing across the Google lineup Listed prices from LiteLLM, OpenRouter and Vercel · last checked Oct 14, 2026
Every weekday
AI moves fast. Here's your debrief.
News, analysis, tools, and more.
For people who build with AI