Gemma 3 27B
Gemma 3 introduces multimodality, supporting vision-language input and text outputs.
text + image → text · made by Google
Sold by 22 ways
| Seller | Lane | Rate |
|---|---|---|
| AWS Bedrock cloud | batch | $0.12per Mtok in$0.19per Mtok outraw.githubusercontent.com · read 2026-10-31 |
| AWS Bedrock cloud | flex | $0.12per Mtok in$0.19per Mtok outraw.githubusercontent.com · read 2026-10-31 |
| AWS Bedrock cloud | standard | $0.23per Mtok in$0.38per Mtok outmodels.dev · read 2026-10-31 |
| AWS Bedrock cloud | priority | $0.4per Mtok in$0.67per Mtok outraw.githubusercontent.com · read 2026-10-31 |
| DeepInfra aggregator | standard | $0.08per Mtok in$0.16per Mtok outapi.deepinfra.com · read 2026-08-25 |
| Hugging Face Inference Providers 2 ways | $0.08per Mtok in$0.16per Mtok out | |
| Hugging Face Inference Providers aggregator | standard | $0.08per Mtok in$0.16per Mtok outmodels.dev · read 2026-10-31 |
| Hugging Face Inference Providers aggregator | deepinfra | $0.08per Mtok in$0.16per Mtok outrouter.huggingface.co · read 2026-08-25 |
| Nous Research aggregator | standard | $0.08per Mtok in$0.35per Mtok out$0.04per Mtok cachedinference-api.nousresearch.com · read 2026-10-15 |
| OpenRouter 6 ways | $0.08per Mtok in$0.35per Mtok out$0.04per Mtok cached | |
| OpenRouter aggregator | standard | $0.08per Mtok in$0.35per Mtok out$0.04per Mtok cachedopenrouter.ai · read 2026-08-24 |
| OpenRouter aggregator | deepinfra/fp8 | $0.08per Mtok in$0.16per Mtok outopenrouter.ai · read 2026-08-25 |
| OpenRouter aggregator | nextbit/fp8 | $0.08per Mtok in$0.35per Mtok out$0.04per Mtok cachedopenrouter.ai · read 2026-10-15 |
| OpenRouter aggregator | parasail/fp8 | $0.08per Mtok in$0.45per Mtok out$0.04per Mtok cachedopenrouter.ai · read 2026-10-18 |
| OpenRouter aggregator | nebius/fp8 | $0.1per Mtok in$0.3per Mtok out$0.04per Mtok cachedopenrouter.ai · read 2026-08-25 |
| OpenRouter aggregator | novita/bf16 | $0.12per Mtok in$0.2per Mtok outopenrouter.ai · read 2026-08-25 |
| OpenRouter aggregatornot on the seller's list today | phala | $0.15per Mtok in$0.46per Mtok out$0.075per Mtok cachedopenrouter.ai · read 2026-08-25 |
| Inference.net aggregator | standard | $0.09per Mtok in$0.25per Mtok outapi.inference.net · read 2026-10-31 |
| Featherless AI aggregator | standard | $0.1per Mtok in$0.3per Mtok outapi.featherless.ai · read 2026-09-20 |
| Nebius Token Factory aggregator | standard | $0.1per Mtok in$0.3per Mtok outtokenfactory.nebius.com · read 2026-08-25 |
| Requesty aggregator | nebius | $0.1per Mtok in$0.3per Mtok out$0.1per Mtok cachedrouter.requesty.ai · read 2026-08-28 |
| Novita AI aggregator | standard | $0.12per Mtok in$0.2per Mtok outapi.novita.ai · read 2026-08-25 |
| ElectronHub aggregator | standard | $0.8per Mtok in$0.8per Mtok outapi.electronhub.ai · read 2026-10-31 |
| Fireworks AI aggregator | standard | $0.9per Mtok in$0.9per Mtok outraw.githubusercontent.com · read 2026-10-31 |
Measured 16 standings
| Place | Board | Metric | Score |
|---|---|---|---|
| 23rdof 49 | Lech mazur writing — Epoch AI | Mean score | 7.99 |
| 31stof 54 | Live bench — Epoch AI | Global average | 49.99 |
| 36thof 111 | Vectara Hallucination Leaderboard | Hallucination Rate | 7.4 |
| 41stof 108 | Math level 5 — Epoch AI | mean_score | 74.037 |
| 68thof 69 | Aider polyglot coding leaderboard | Percent correct | 4.9 |
| 71stof 72 | Aider polyglot — Epoch AI | Percent correct | 4.9 |
| 111thof 165 | LMArena · Vision | Score (Elo) | 1160 |
| 131stof 131 | Artificial Analysis · Agentic Index | Index | 0.1 |
| 155thof 163 | Artificial Analysis · Coding Index | Index | 10.1 |
| 180thof 197 | Lmca — Epoch AI | Score | 12.28 |
| 198thof 414 | LMArena · Text | Score (Elo) | 1365 |
| 210thof 260 | Critpt — Epoch AI | Accuracy | 0 |
| 211thof 235 | Dtbench — Epoch AI | Accuracy | 52.53 |
| 229thof 235 | Scicode — Epoch AI | Score | 23.264 |
| 242ndof 306 | OTIS Mock AIME 2024-2025 | mean_score | 22.5 |
| 252ndof 326 | GPQA Diamond | mean_score | 47.727 |
About
Gemma 3 27B — a text model from Google, sold by 12 companies from $0.08 in and $0.16 out per million tokens, placed 36th of 111 on Vectara Hallucination Leaderboard.
It takes text and images and returns text, with a context window of 32,768 tokens. It was published in March 2025, trained on material up to August 2024. Its sellers say it can call a tool. The catalogue files it under reasoning and code. Twelve companies sell it. The cheapest is $0.08 in and $0.16 out per million tokens at DeepInfra. Beside the standard rate there are batch, flex, faster, separately routed and regional lanes. It has been measured on 16 boards, and stands best at 36th of 111 on Vectara Hallucination Leaderboard.
Every current figure
- Maker
- Register
- model
- Takes
- text + image
- Returns
- text
- Context
- 32,768 tokens
- Published
- March 2025
- Knowledge to
- August 2024
- Parameters
- 27 billion
- Licence
- not read
- Sellers
- 12
- Maker's own price
- not read
- Price
- $0.08 in and $0.16 out per million tokens — DeepInfra
- Boards
- 16
- Best place
- 36th of 111 — Vectara Hallucination Leaderboard
Known as 5 names
gemma-3-27b-itgoogle/gemma-3-27b-itcontext-labs/google-gemma-3-27b-itunsloth/gemma-3-27b-itnebius/google/gemma-3-27b-it