Skip to content
Weights Not Walls
Menu

google

gemma-2-9b-it

Reviewedgated on the Hub
Size
9.2B parameterscounted from the safetensors header on the Hub
Licence
unknown, Gemma Terms of UseOur reading, check the licence: Commercial use is allowed with no user or revenue cap, but the Gemma Prohibited Use Policy (a separate document, not in this text) is part of the terms, and you must pass the use restrictions on as enforceable terms to anyone you distribute to. Models distilled from Gemma or trained on its outputs to perform like it count as Model Derivatives and stay under these terms, and Google may restrict use remotely if it believes the terms are broken. The page says these terms cover the models in its appendix (Gemma 1 to 3, 3n and variants) and that Gemma 4 has a separate licence.
On the Hub since
24 Jun 2024the date the repo was created
Context length
Not known
Fine-tune
of google/gemma-2-9bGemma2ForCausalLM
On Hugging Face
google/gemma-2-9b-it163k downloads in the last 30 days, 1.2k likes

Scores

5 independent results on 5 eval sets
GPQA diamondRe-run by Epoch AI
  • default setting27.5%±2.327 Jan 2025, log

Epoch's name: gemma-2-9b-it. Best closed model on this set: gpt-6.2-astra_max, 96.8% (OpenAI). All 341 results

MATH level 5Re-run by Epoch AI
  • default setting21.0%±0.927 Jan 2025, log

Epoch's name: gemma-2-9b-it. Best closed model on this set: gpt-5-2025-08-07_high, 98.1% (OpenAI). All 108 results

OTIS Mock AIME 2024-2025Re-run by Epoch AI
  • default setting0.6%±0.47 Mar 2025, log

Epoch's name: gemma-2-9b-it. Best closed model on this set: claude-fable-5_high, 100.0% (Anthropic). All 318 results

LiveBenchReported

Epoch's name: gemma-2-9b-it. Best closed model on this set: gemini-2.5-pro-exp-03-25, 82.3% (Google DeepMind). All 54 results

MMLUReported

Epoch's name: gemma-2-9b-it. Best closed model on this set: gpt-4o-2024-11-20, 88.1% (OpenAI). All 217 results

Claimed by the maker Claimed

The model card has no claimed scores in a machine-readable form (the card's model-index), so none are shown. The maker may give scores in the card text: read the card.

Other views, which we link to and do not copy: look this model up on Artificial Analysis and the Arena leaderboard.

Independent scores from Epoch AI (CC-BY 4.0), fetched 19 Nov 2026: Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from ‘https://epoch.ai/benchmarks’ [online resource]. Each row links its own source. What the badges mean.

Run it yourself

file sizes read from the Hub 19 Nov 2026

Memory needed = weights + context cache + 0.50 GB for the runtime. Context cache at 8k tokens: 2.6 GB, worked out from the model's config (hugging-quants/gemma-2-9b-it-AWQ-INT4), at 16-bit precision. Choose a machine above to see a verdict for each format. Method

GGUF for llama.cpp, Ollama, LM Studio and others

bartowski/gemma-2-9b-it-GGUF (converted by bartowski), 32k downloads. Sizes are the real files on the Hub, all parts added up.

FormatSizeMemory neededGet it
IQ3_XXS3.5 GBfile size6.7 GB3.5 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:IQ3_XXS
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:IQ3_XXS
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-IQ3_XXS.gguf" --local-dir gemma-2-9b-it-GGUF
IQ3_XS3.9 GBfile size7.0 GB3.9 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:IQ3_XS
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:IQ3_XS
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-IQ3_XS.gguf" --local-dir gemma-2-9b-it-GGUF
Q3_K_S4.0 GBfile size7.2 GB4.0 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:Q3_K_S
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:Q3_K_S
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-Q3_K_S.gguf" --local-dir gemma-2-9b-it-GGUF
IQ3_M4.2 GBfile size7.3 GB4.2 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:IQ3_M
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:IQ3_M
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-IQ3_M.gguf" --local-dir gemma-2-9b-it-GGUF
Q3_K_M4.4 GBfile size7.6 GB4.4 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:Q3_K_M
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:Q3_K_M
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-Q3_K_M.gguf" --local-dir gemma-2-9b-it-GGUF
Q3_K_L4.8 GBfile size7.9 GB4.8 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:Q3_K_L
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:Q3_K_L
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-Q3_K_L.gguf" --local-dir gemma-2-9b-it-GGUF
IQ4_XS4.8 GBfile size8.0 GB4.8 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:IQ4_XS
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:IQ4_XS
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-IQ4_XS.gguf" --local-dir gemma-2-9b-it-GGUF
Q3_K_XL5.0 GBfile size8.1 GB5.0 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:Q3_K_XL
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:Q3_K_XL
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-Q3_K_XL.gguf" --local-dir gemma-2-9b-it-GGUF
Q4_K_S5.1 GBfile size8.2 GB5.1 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:Q4_K_S
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:Q4_K_S
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-Q4_K_S.gguf" --local-dir gemma-2-9b-it-GGUF
Q4_K_M5.4 GBfile size8.5 GB5.4 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:Q4_K_M
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:Q4_K_M
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-Q4_K_M.gguf" --local-dir gemma-2-9b-it-GGUF
Q4_K_L5.6 GBfile size8.7 GB5.6 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:Q4_K_L
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:Q4_K_L
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-Q4_K_L.gguf" --local-dir gemma-2-9b-it-GGUF
Q5_K_S6.0 GBfile size9.2 GB6.0 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:Q5_K_S
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:Q5_K_S
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-Q5_K_S.gguf" --local-dir gemma-2-9b-it-GGUF
Q5_K_M6.2 GBfile size9.3 GB6.2 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:Q5_K_M
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:Q5_K_M
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-Q5_K_M.gguf" --local-dir gemma-2-9b-it-GGUF
Q5_K_L6.4 GBfile size9.5 GB6.4 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:Q5_K_L
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:Q5_K_L
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-Q5_K_L.gguf" --local-dir gemma-2-9b-it-GGUF
Q6_K7.1 GBfile size10.2 GB7.1 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:Q6_K
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:Q6_K
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-Q6_K.gguf" --local-dir gemma-2-9b-it-GGUF
Q6_K_L7.3 GBfile size10.4 GB7.3 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:Q6_K_L
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:Q6_K_L
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-Q6_K_L.gguf" --local-dir gemma-2-9b-it-GGUF
Q87.3 GBfile size10.4 GB7.3 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:Q8
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:Q8
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-Q6_K-Q8.gguf" --local-dir gemma-2-9b-it-GGUF
Q8_09.2 GBfile size12.3 GB9.2 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:Q8_0
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:Q8_0
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-Q8_0.gguf" --local-dir gemma-2-9b-it-GGUF
F1610.0 GBfile size13.1 GB10.0 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:F16
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:F16
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-Q8_0-f16.gguf" --local-dir gemma-2-9b-it-GGUF
Q8_0_L10.0 GBfile size13.1 GB10.0 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:Q8_0_L
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:Q8_0_L
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-Q8_0_L.gguf" --local-dir gemma-2-9b-it-GGUF
F3234.4 GBfile size37.6 GB34.4 + 2.6 + 0.50
Commands
llama.cpp
llama-server -hf bartowski/gemma-2-9b-it-GGUF:F32
Ollama
ollama run hf.co/bartowski/gemma-2-9b-it-GGUF:F32
Download
hf download bartowski/gemma-2-9b-it-GGUF --include "gemma-2-9b-it-f32.gguf" --local-dir gemma-2-9b-it-GGUF

MLX for Apple silicon

FormatSizeMemory neededRepo
MLXTheBlueObserver/gemma-2-9b-it-MLX-104ce3.8 GBfile size6.9 GB3.8 + 2.6 + 0.50Download page
MLX 6bitmlx-community/gemma-2-9b-it-6bit7.0 GBfile size10.1 GB7.0 + 2.6 + 0.50Download page

Original weights size from the safetensors header, for vLLM, SGLang, Transformers

FormatSizeMemory neededRepo
BF16original weights17.2 GBfile size20.3 GB17.2 + 2.6 + 0.50
hf download google/gemma-2-9b-it --local-dir gemma-2-9b-it
Other conversions (9; sizes not read)

The Hub tags 100 or more quantised repos as conversions of this model; we read the most downloaded and show the maker's own first.

Run it by API

Hugging Face Inference Providers, fetched 19 Nov 2026, 13:58 UTC

No provider on Hugging Face Inference Providers serves this model (as of 19 Nov 2026, 13:58 UTC). Other hosts may; we only list what the router reports.

Not built yet: one key for every open model, with BlockRun, and our own tested-quality score per provider. Until then, precision and measured speed are the best guide to a provider here.

Planned for this page Not built yet

Try it in private chat
A link to Solana AI's confidential chat for the models it serves. This model is not listed as served there.
Fine-tune it
Fine-tuning a few base models from this page and publishing the result here.
Creators and earnings
Pages for people who publish fine-tunes and adapters of this model, and what they earn.
Fund and bounties
Bounties for new open models, paid from a public fund. How the fund will work.
Contribute compute
How to give compute to open model work, with Akash.

Model data from the Hugging Face Hub 19 Nov 2026, 13:43 UTC. The repo. Found a mistake on this page? Tell us.