Skip to content
Weights Not Walls
Menu

google

gemma-3-27b-it

Reviewedgated on the Hubaccepts images
Size
27.4B 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
1 Mar 2025the date the repo was created
Context length
128k tokensas an API provider reports it
Fine-tune
of google/gemma-3-27b-ptGemma3ForConditionalGeneration
On Hugging Face
google/gemma-3-27b-it291k downloads in the last 30 days, 2.1k likes

Scores

11 independent results on 11 eval sets
GPQA diamondRe-run by Epoch AI
  • default setting47.7%±2.727 Aug 2026, log

Epoch's name: gemma-3-27b-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 setting74.0%±1.113 Mar 2025, log

Epoch's name: gemma-3-27b-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 setting22.5%±5.327 Aug 2026, log

Epoch's name: gemma-3-27b-it. Best closed model on this set: claude-fable-5_high, 100.0% (Anthropic). All 318 results

Epoch's name: gemma-3-27b-it. Best closed model on this set: gpt-5-2025-08-07_high, 88.0% (OpenAI). All 72 results

CritPtReported

Epoch's name: gemma-3-27b-it. Best closed model on this set: gemini-4-carbon_high, 36.0% (Google DeepMind). All 271 results

DTBenchReported

Epoch's name: gemma-3-27b-it. Best closed model on this set: claude-fable-5-2_unknown, 99.2% (Anthropic). All 241 results

Epoch's name: chutes/Gemma-3-27b-It. Best closed model on this set: gpt-5-2025-08-07_medium, 97.2% (OpenAI). All 61 results

GeoBenchReported

Epoch's name: gemma-3-27b-it. Best closed model on this set: claude-fable-5_high, 89.0% (Anthropic). All 51 results

Epoch's name: gemma-3-27b-it. Best closed model on this set: gpt-5-2025-08-07_medium, 86.0% (OpenAI). All 49 results

LiveBenchReported

Epoch's name: gemma-3-27b-it. Best closed model on this set: gemini-2.5-pro-exp-03-25, 82.3% (Google DeepMind). All 54 results

LMCAReported

Epoch's name: gemma-3-27b-it. Best closed model on this set: claude-fable-5-2_xhigh, 70.6% (Anthropic). All 203 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: 3.9 GB, worked out from the model's config (gaunernst/gemma-3-27b-it-int4-awq), 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

google/gemma-3-27b-it-qat-q4_0-gguf (the maker's own conversion), 210 downloads. Sizes are the real files on the Hub, all parts added up. For image input you also need the projector file mmproj-model-f16-27B.gguf (0.80 GB), not counted below.

FormatSizeMemory neededGet it
Q4_016.0 GBfile size20.4 GB16.0 + 3.9 + 0.50
Commands
llama.cpp
llama-server -hf google/gemma-3-27b-it-qat-q4_0-gguf:Q4_0
Ollama
ollama run hf.co/google/gemma-3-27b-it-qat-q4_0-gguf:Q4_0
Download
hf download google/gemma-3-27b-it-qat-q4_0-gguf --include "gemma-3-27b-it-q4_0.gguf" --local-dir gemma-3-27b-it-qat-q4_0-gguf

MLX for Apple silicon

FormatSizeMemory neededRepo
MLXstaccs/gemma-3-27b-it-lecore-mlx11.0 GBfile size15.4 GB11.0 + 3.9 + 0.50Download page
MLX 4bitOutlier-Ai/gemma-3-27b-it-MLX-4bit14.2 GBfile size18.5 GB14.2 + 3.9 + 0.50Download page

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

FormatSizeMemory neededRepo
BF16original weights51.1 GBfile size55.5 GB51.1 + 3.9 + 0.50
hf download google/gemma-3-27b-it --local-dir gemma-3-27b-it
Other conversions (18; 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

5 providers serve this model through the Hugging Face router. Prices are US dollars per million tokens. Precision is what the provider reports; a lower precision than the original weights can change answers. Live first, then cheapest output.

ProviderInOutContextPrecisionTokens/sFirst tokenUptime 24hToolsJSON
DeepInfra$0.08$0.16128kfp8241.6 s99.7%yesyes
Nebius$0.10$0.30107kfp8361.2 s99.7%noyes
NextBit$0.08$0.35128kfp8180.7 s99.4%noyes
Parasail$0.08$0.45128kfp8330.7 s100.0%noyes
featherless-ai???not given???not givennot given

Through the router, with a Hugging Face token (the router bills your Hugging Face account). Change the name after the colon to pick another provider.

curl https://router.huggingface.co/v1/chat/completions \ -H "Authorization: Bearer $HF_TOKEN" -H "Content-Type: application/json" \ -d '{"model": "google/gemma-3-27b-it:deepinfra", "messages": [{"role": "user", "content": "Hello"}]}'

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.

Fine-tunes of this model

from our snapshot of the Hub

Repos that declare this model as their base, most downloaded first. Listing is not a recommendation; unreviewed ones have not been checked.

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.