Gemma 3 Base: pricing, context window and benchmarks
Gemma 3 Base is a open-weight model from Google DeepMind. It offers a 65,536-token context window, costs $3.09 per million input tokens and $10.93 per million output tokens, and averages 71.1 across the index's benchmark suite.
Pricing
Gemma 3 Base costs $3.09 per million input tokens and $10.93 per million output tokens. A blended 3:1 input-to-output mix works out to about $5.05 per million tokens. Cached input is listed at $0.43 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $3.09 | per 1M tokens |
| Output | $10.93 | per 1M tokens |
| Cached input | $0.43 | per 1M tokens |
| Blended 3:1 | $5.05 | estimate per 1M tokens |
| 1M in + 1M out | $14.02 | estimate |
Benchmarks
Scores below are modelled estimates on a 0–100 scale. Use them to compare relative position, not as official vendor results.
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 75.2 | 327 | 46th | |
| GPQA Diamond | 64.7 | 365 | 39th | |
| SWE-bench Verified | 60.6 | 315 | 48th | |
| HumanEval | 77.6 | 297 | 51th | |
| MATH-500 | 72.8 | 359 | 40th | |
| AIME 2025 | 64.3 | 352 | 41th | |
| LiveCodeBench | 62.4 | 345 | 43th | |
| IFEval | 79.7 | 338 | 44th | |
| TruthfulQA | 75.6 | 318 | 47th | |
| HellaSwag | 81.6 | 358 | 40th | |
| MMMU | 67.7 | 349 | 42th | |
| Arena Elo | 1350.7 | 362 | 40th |
What Gemma 3 Base is good for
Gemma 3 Base is an open-weight model, so it can be self-hosted or run through a third-party inference provider. That matters for data residency, fine-tuning and long-run cost control. At $3.09 per million input tokens (hosted list price) and a 65,536-token context window, it is a practical option for teams that want more control than a closed API allows.
Strengths
- Input price of $3.09 per million tokens is at or below the index median.
- Open weights make self-hosting, fine-tuning and data-residency review possible.
- Accepts more than text, including image input where the provider supports it.
Trade-offs
- Average benchmark score is below the index median; test it on your own task.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Fusion | OpenRouter | 16384 | $8.41 | 71.1 |
| Gemma 3 Base | Google DeepMind | 65536 | $3.09 | 71.1 |
| Llama 3 Hyperbolic Base | Hyperbolic | 32768 | $3.02 | 71.1 |
| Mixtral Base | Mistral AI | 65536 | $2.95 | 71.1 |
| Command A Mini | Cohere | 8192 | $8.12 | 71.0 |
| OLMo 2 | Allen Institute for AI | 65536 | $2.99 | 71.0 |
FAQ
How much does Gemma 3 Base cost?
Gemma 3 Base is listed at $3.09 per million input tokens and $10.93 per million output tokens. Cached input is $0.43 per million tokens. A blended 3:1 input-to-output mix is about $5.05 per million tokens.
What is the context window of Gemma 3 Base?
Gemma 3 Base has a context window of 65,536 tokens and can emit up to 4,096 output tokens per request in this index.
Is Gemma 3 Base open weights?
Yes. Gemma 3 Base is listed as an open-weight model under a open weights licence.
How does Gemma 3 Base score on benchmarks?
Gemma 3 Base averages 71.1 across the scored benchmarks in this index, ranking #344 of 600 models. Its strongest relative result is Arena Elo at 1350.7.
What is Gemma 3 Base best for?
Gemma 3 Base is an open-weight model, so it can be self-hosted or run through a third-party inference provider. That matters for data residency, fine-tuning and long-run cost control. At $3.09 per million input tokens (hosted list price) and a 65,536-token context window, it is a practical option for teams that want more control than a closed API allows.