Gemma 3 4B Instruct
GoogleGemmaOpen WeightGemma · Commercial OK
Description
Gemma 3 4B is a 4-billion-parameter vision-language model from Google, handling text and image input and generating text output. It features a 128K context window, multilingual support, and open weights. Suitable for question answering, summarization, reasoning, and image understanding tasks.
Release Date
2025-03-12
Parameters
4.0B
Context Length
131K
Modalities
image, text
Capability Radar
14
general
10
coding
22
reasoning
17
scienceest.
18
agents
70
multimodal
Science uses a reasoning proxy when dedicated science benchmarks are unavailable.
Rankings
| Domain | #Rank | Score | Source |
|---|---|---|---|
| Code Ranking | 464 | 7.0 | AA |
| General Ranking | 489 | 14.0 | AA |
| Math Reasoning | 290 | 24.0 | AA |
| Multimodal Ranking | 76 | 65.0 | LS |
| Reasoning | 101 | 36.0 | LS |
| Science | 479 | 14.0 | AA |
Benchmark Scores (LLM Stats)
Biology
GPQA
30.8%SR
Code
HumanEval
71.3%SR
LiveCodeBench
12.6%SR
Factuality
FACTS Grounding
70.1%SR
SimpleQA
4.0%SR
Finance
MMLU-Pro
43.6%SR
General
IFEval
90.2%SR
Natural2Code
70.3%SR
MBPP
0.63 / 100SR
Global-MMLU-Lite
54.5%SR
MMMU (val)
48.8%SR
BIG-Bench Extra Hard
11.0%SR
Image To Text
DocVQA
75.8%SR
VQAv2 (val)
62.4%SR
TextVQA
57.8%SR
Language
BIG-Bench Hard
72.2%SR
WMT24++
46.8%SR
ECLeKTic
4.6%SR
Math
GSM8k
89.2%SR
MATH
75.6%SR
MathVista-Mini
50.0%SR
HiddenMath
43.0%SR
Multimodal
AI2D
74.8%SR
ChartQA
68.8%SR
InfoVQA
50.0%SR
Reasoning
Bird-SQL (dev)
36.3%SR
AA Evaluation Indices
Math Index12.7
Intelligence Index1.1
Math 5000.8
Mmlu Pro0.4
Gpqa0.3
Ifbench0.3
Aime 250.1
Livecodebench0.1
Scicode0.1
Aime0.1
Lcr0.1
Hle0.1
Tau20.0
Terminalbench Hard0.0
LLM Stats Category Scores
Instruction Following90
Structured Output90
Image To Text70
Grounding70
Math60
Multimodal60
Vision60
Reasoning50
Healthcare50
Language40
Legal40
Factuality40
Finance40
General40
Code40
Physics30
Biology30
Chemistry30
Pricing
Input Price$0.04 / 1M tokens
Output Price$0.08 / 1M tokens
Blended Price (3:1)$0.05 / 1M tokens
Speed
Tokens/sec0.0
Time to First Token0.00s
Time to Answer0.00s
Provider Price Ranking
Provider Price Ranking
6 providers
Cheapest: ChutesMost Expensive: NanoGPT
ProviderInputOutput
1ChutesCheapest
$0.01
$0.0272
2GooglePRIMARY
$0.04
$0.08
3Kilo Gateway
$0.04
$0.08
4Amazon Bedrock
$0.04
$0.08
5OpenRouter
$0.05
$0.1
6NanoGPT
$0.2006
$0.2006
Compare pricing across different API providers for this model.