Gemma 3 4B
Gemma 3 4BRankingconfiguration unstated?
Alternatives?
CheaperGemma 2 9B ItOpenwins 48%-12%$0.03 / $0.06Context: 131K → 8K+2
BetterGPT-5 Minihighwins 57%+1,518%$0.053OpenGemma 3 12Bunstatedwins 58%+47%$0.05 / $0.10Memory (4-bit): 2.2 GB → 6.1 GBCheaperGPT OSS 20BunstatedOpenwins 45%-15%$0.018 / $0.09Memory (4-bit): 2.2 GB → 13 GB+1BetterGPT-6 Lunamaxwins 69%+371%$0.068Function calling only with none · Chat Completions+3Gemma 2 9B ItOpenwins 48%-12%$0.03 / $0.06Context: 131K → 8K+2GPT OSS 20BunstatedOpenwins 45%-15%$0.018 / $0.09Memory (4-bit): 2.2 GB → 13 GB+1GPT-5 Minihighwins 57%+1,518%$0.053GPT-6 Lunamaxwins 69%+371%$0.068Function calling only with none · Chat Completions+3GPT-5.6 Lunaxhighwins 69%+959%$0.085GPT-5.6 Terraxhighwins 71%+10,488%$0.63Prompts with >272K input tokensGPT-6.1 Solmaxwins 74%+9,312%$0.72Remove unsupported parameters+5Gemma 3 12BunstatedOpenwins 58%+47%$0.05 / $0.10Memory (4-bit): 2.2 GB → 6.1 GBGemma 3n E4B InstructunstatedOpenwins 53%——Context: 131K → 32K+1
Granite 4.1 8BunstatedOpensame——Memory (4-bit): 2.2 GB → 4.4 GB+1Gemma 2 9B ItOpenwins 48%-12%$0.03 / $0.06Context: 131K → 8K+2
GLM-4Openwins 45%——Context: not verified+2Gemini 4 Argonhighwins 80%—*$1.99Synchronous output stops at 65,536+2Gemini 3.8 Flashhighwins 77%—*$1.24Sampling parameters are ignored+1Gemini 3.1 Pro Previewunstatedwins 77%—$2 / $12Cannot use both thinking controlsGemini 3.6 Flashhighwins 76%—*$1.60Gemini 3 Flash Previewunstatedwins 74%—$0.50 / $3Gemini 2.5 Prounstatedwins 72%—$1.25 / $10Gemini 3.5 Flash Liteunstatedwins 72%—$0.30 / $2.50Gemma 4 31BunstatedOpenwins 70%—$0.10 / $0.34Memory (4-bit): 2.2 GB → 19 GBAll Google models →Run it yourself
estimateBenchmarks
all categoriesFAQ
How much does Gemma 3 4B cost?›
$0.04 per 1M input tokens and $0.05 per 1M output tokens; cached input $0.02 per 1M (cheapest host: Parasail).
Is Gemma 3 4B free to use?›
There is no free API; the cheapest paid price is $0.04 per 1M input tokens. The weights are open, so you can run it yourself.
How good is Gemma 3 4B, and what is it best at?›
#154 of 196 in Chat & Knowledge, #36 of 39 in Agents & Tool Use, #147 of 195 in Writing & Language, #31 of 41 in Vision & Documents.
Is Gemma 3 4B better than Llama 3.1 Nemotron 70B Instruct?›
On Arena Text · creative writing, Gemma 3 4B scores 1,275 vs 1,276 — about the same.
How can I use Gemma 3 4B?›
Through the API of AWS Bedrock, DeepInfra, Parasail.
What is the context window of Gemma 3 4B?›
131K tokens, with up to 16K output tokens.
When was Gemma 3 4B released? Is there a newer version?›
Released 2025-02-20. It is the current version in its line.
Is Gemma 3 4B open source?›
Yes, the weights are open (gemma licence).
Can I run Gemma 3 4B locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Gemma 3 4B read images?›
It accepts text and images as input.
What is Gemma 3 4B?›
Gemma 3 4B is an open-weights language model from Google, released Feb 2025.
What are the best alternatives to Gemma 3 4B?›
Gemma 2 9B It (cheaper), GPT-5 Mini (better), Gemma 3 12B (open weights).