Gemma 3 4B
Gemma 3 introduces multimodality, supporting vision-language input and text outputs.
text + image → text · made by Google
Sold by 14 ways
| Seller | Lane | Rate |
|---|---|---|
| AWS Bedrock cloud | batch | $0.02per Mtok in$0.04per Mtok outraw.githubusercontent.com · read 2026-10-31 |
| AWS Bedrock cloud | flex | $0.02per Mtok in$0.04per Mtok outraw.githubusercontent.com · read 2026-10-31 |
| AWS Bedrock cloud | standard | $0.04per Mtok in$0.08per Mtok outmodels.dev · read 2026-10-31 |
| AWS Bedrock cloud | priority | $0.07per Mtok in$0.14per Mtok outraw.githubusercontent.com · read 2026-10-31 |
| OpenRouter 3 ways | $0.05per Mtok in$0.1per Mtok out$0.02per Mtok cached | |
| OpenRouter aggregator | parasail/bf16 | $0.04per Mtok in$0.05per Mtok out$0.02per Mtok cachedopenrouter.ai · read 2026-10-16 |
| OpenRouter aggregatornot on the seller's list today | parasail/fp8 | $0.04per Mtok in$0.05per Mtok out$0.02per Mtok cachedopenrouter.ai · read 2026-10-15 |
| OpenRouter aggregator | standard | $0.05per Mtok in$0.1per Mtok out$0.02per Mtok cachedopenrouter.ai · read 2026-10-28 |
| OpenRouter aggregator | deepinfra/bf16 | $0.05per Mtok in$0.1per Mtok outopenrouter.ai · read 2026-10-20 |
| DeepInfra aggregator | standard | $0.05per Mtok in$0.1per Mtok outapi.deepinfra.com · read 2026-08-25 |
| Featherless AI aggregator | standard | $0.05per Mtok in$0.1per Mtok outapi.featherless.ai · read 2026-09-20 |
| Hugging Face Inference Providers 2 ways | $0.05per Mtok in$0.1per Mtok out | |
| Hugging Face Inference Providers aggregator | standard | $0.05per Mtok in$0.1per Mtok outmodels.dev · read 2026-10-31 |
| Hugging Face Inference Providers aggregator | deepinfra | $0.05per Mtok in$0.1per Mtok outrouter.huggingface.co · read 2026-08-25 |
| Inference.net aggregator | standard | $0.05per Mtok in$0.1per Mtok outapi.inference.net · read 2026-10-31 |
| ElectronHub aggregator | standard | $0.15per Mtok in$0.25per Mtok outapi.electronhub.ai · read 2026-10-31 |
Measured 7 standings
| Place | Board | Metric | Score |
|---|---|---|---|
| 29thof 111 | Vectara Hallucination Leaderboard | Hallucination Rate | 6.4 |
| 163rdof 163 | Artificial Analysis · Coding Index | Index | 2.7 |
| 197thof 197 | Lmca — Epoch AI | Score | 2.78 |
| 218thof 235 | Dtbench — Epoch AI | Accuracy | 50.93 |
| 261stof 306 | OTIS Mock AIME 2024-2025 | mean_score | 7.5 |
| 282ndof 414 | LMArena · Text | Score (Elo) | 1303 |
| 320thof 326 | GPQA Diamond | mean_score | 23.232 |
About
Gemma 3 4B — a text model from Google, sold by 7 companies from $0.04 in and $0.08 out per million tokens, placed 29th 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 December 2024, trained on material up to August 2024. The catalogue files it under reasoning. Seven companies sell it. The cheapest is $0.04 in and $0.08 out per million tokens at AWS Bedrock. Beside the standard rate there are batch, flex, faster, separately routed and regional lanes. It has been measured on 7 boards, and stands best at 29th of 111 on Vectara Hallucination Leaderboard.
Every current figure
- Maker
- Register
- model
- Takes
- text + image
- Returns
- text
- Context
- 32,768 tokens
- Published
- December 2024
- Knowledge to
- August 2024
- Parameters
- 4.3 billion
- Licence
- not read
- Sellers
- 7
- Maker's own price
- not read
- Price
- $0.04 in and $0.08 out per million tokens — AWS Bedrock
- Boards
- 7
- Best place
- 29th of 111 — Vectara Hallucination Leaderboard
Known as 3 names
google/gemma-3-4b-ittbmod/gemma-3-4b-itunsloth/gemma-3-4b-it