Skip to main content
Back to Explore

Gemma 3 4B Instruct

Google·Released 2025-03-12
Open SourceMultimodal

Pricing

Input

$0.04

per 1M tokens

Output

$0.08

per 1M tokens

Blended

$0.05

per 1M tokens

Cheaper than 90% of models. Median price is $0.54/1M tokens.

Cost Calculator

Tokens per day1M
100K100M

Daily

$0.05

Monthly

$1.50

vs. Similar Models

Llama 3.2 Instruct 1BQ:0.0
$0.050%
Gemma 3n E4B InstructQ:+0.1
$0.03-50%
Llama 3 Instruct 8BQ:+0.1
$0.07+40%
Apertus 8B InstructQ:-0.1
$0.13+150%

Performance

34

tokens/sec

Faster than 3% of models

1.19

seconds

Faster than 44% of models

1.19

seconds

Faster than 63% of models

Market Median

94 tok/s

64% slower

Median TTFT

1.11s

7% slower

Throughput/Dollar

685

tok/s per $/1M

Speed Comparison

OpenAI: o3 Pro
34 tok/s-0%
Claude 4 Opus (Reasoning)
34 tok/s+0%
Llama 3.1 Instruct 70B
35 tok/s+2%

Benchmarks

MMLU-Pro
41.7%
GPQA Diamond
29.1%
HLE
5.2%
LiveCodeBench
11.2%
SciCode
7.3%
TerminalBench Hard
0.8%
MATH-500
76.6%
AIME
6.3%
AIME 2025
12.7%
IFBench
28.3%
Long Context Recall
5.7%
Tau2
5.0%
Market AverageTop Score

Open Source

Quick Compare

Similar Models

Compare all 7 models