New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Gemma 2 9B
About Gemma 2 9B
Gemma 2 9B is an AI model developed by Google DeepMind, in the language category, released in 2024, made available as an open-weights model with 9B params.
On this page you'll find Gemma 2 9B's full specifications, including 35 benchmark results. Review provider pricing and benchmark scores below, or compare Gemma 2 9B head-to-head with other language models.
Links
Run Gemma 2 9B locally — what to buy
Share this build ↗Sized to what Gemma 2 9B actually needs (~8k context at Q4), not to the biggest GPU: Good = cheapest that runs it, Better = best value, Best = most headroom. Speed is a hardware estimate; anything past ~120 tok/s is shown as instant.
Software support
✓ measured · · compatible · — not supported. Informational only — speed is hardware-based.
Community
No reports yet. Own this model on some hardware? Be the first to confirm it runs.
No comments yet. Start the discussion.
Gemma 2 9B benchmark scores▶
| Benchmark | Category | Score | Variant |
|---|---|---|---|
| HumanEval | code | 71.7% | instruct-pass@10-cited-tulu3 |
| HumanEval | code | 61% | instruct-cited-phi35moe |
| HumanEval+ | code | 67% | instruct-pass@10-cited-tulu3 |
| MBPP | code | 69.3% | instruct-cited-phi35moe |
| AlpacaEval 2 | general | 43.7% | instruct-lc-win-cited-tulu3 |
| ArenaHard | general | 43.6% | instruct-cited-exaone35 |
| ArenaHard | general | 42% | instruct-cited-phi35moe |
| IFEval | general | 69.9% | instruct-cited-tulu3 |
| LiveBench | general | 32.1 | instruct-cited-exaone35 |
| MMLU | general | 74.6% | instruct-cited-tulu3 |
| MMLU | general | 70.6% | base-cited-olmo2 |
| MMLU-Pro | general | 50.1% | instruct-cited-phi35moe |
| MMLU-Pro | general | 42% | base-cited-olmo2 |
| MMMLU | general | 63.8% | instruct-cited-phi35moe |
| AGIEval | knowledge | 57.3 | base-cited-olmo2 |
| ARC-Challenge | knowledge | 89.8 | instruct-cited-phi35moe |
| ARC-Challenge | knowledge | 89.5 | base-cited-olmo2 |
| TriviaQA | knowledge | 81.8 | base-cited-olmo2 |
| GSM8K | math | 84.9% | instruct-cited-phi35moe |
| GSM8K | math | 79.7% | instruct-cited-tulu3 |
| GSM8K | math | 70.1% | base-cited-olmo2 |
| MATH | math | 50.9% | instruct-cited-phi35moe |
| MATH | math | 29.8% | instruct-cited-tulu3 |
| MGSM | math | 75.1 | instruct-cited-phi35moe |
| BIG-Bench Hard | reasoning | 63.5% | instruct-cited-phi35moe |
| DROP | reasoning | 63% | base-cited-olmo2 |
| DROP | reasoning | 58.8% | instruct-cited-tulu3 |
| GPQA Diamond | reasoning | 29.2% | instruct-cited-phi35moe |
| HellaSwag | reasoning | 87.3 | base-cited-olmo2 |
| HellaSwag | reasoning | 80.9 | instruct-cited-phi35moe |
| PIQA | reasoning | 83.7 | instruct-cited-phi35moe |
| Winogrande | reasoning | 78.8 | base-cited-olmo2 |
| Winogrande | reasoning | 74 | instruct-cited-phi35moe |
| TruthfulQA | safety | 76.6% | instruct-cited-phi35moe |
| TruthfulQA | safety | 61.4% | instruct-cited-tulu3 |
More models like Gemma 2 9B
Frequently asked questions
What is Gemma 2 9B? ▶
Gemma 2 9B is an AI model developed by Google DeepMind, in the language category, released in 2024. It is tracked on GenAIList with its specifications, benchmark scores and provider pricing.
Who created Gemma 2 9B? ▶
Gemma 2 9B was developed by Google DeepMind and released in 2024.
Is Gemma 2 9B open source or proprietary? ▶
Gemma 2 9B ships as an open-weights model: you can download and self-host the weights, though the licence may place some restrictions on use.
How much does Gemma 2 9B cost? ▶
Pricing for Gemma 2 9B depends on the provider. See the providers table on this page for the latest API rates.
How does Gemma 2 9B perform on benchmarks? ▶
Gemma 2 9B is benchmarked across 35 evaluations on GenAIList, including HumanEval (71.7%). See the full benchmark table below and compare it with other models.
No videos yet. Know a good one? Add it below.
Know a great video about Gemma 2 9B?
Add YouTube Video
Add Tweet/Discussion
Report an Issue
Help us maintain accurate information in our database.
Suggest an Edit
Tell us what needs to be changed or corrected for