Skip to main content
Back to Explore

Gemma 4 E4B (Reasoning)

Google·Released 2026-04-03
Open SourceMultimodal

Pricing

Input

$0.30

per 1M tokens

Output

$1.25

per 1M tokens

Blended

$0.54

per 1M tokens

Cheaper than 53% of models. Median price is $0.67/1M tokens.

Cost Calculator

Tokens per day1M
100K100M

Daily

$0.54

Monthly

$16.11

vs. Similar Models

Mistral: Mistral Medium 3Q:0.0
$0.80+49%
Devstral MediumQ:-0.1
$0.80+49%
Llama Nemotron Super 49B v1.5 (Reasoning)Q:-0.1
$0.40-26%
Mistral Small 4 (Non-reasoning)Q:-0.1
$0.26-51%

Performance

60

tokens/sec

Faster than 29% of models

3.39

seconds

Faster than 21% of models

36.62

seconds

Faster than 14% of models

Market Median

93 tok/s

35% slower

Median TTFT

1.13s

201% slower

Throughput/Dollar

112

tok/s per $/1M

Speed Comparison

DeepSeek V4 Pro (Reasoning, High Effort)
60 tok/s+0%
Qwen3 Max Thinking (Preview)
60 tok/s-0%
Qwen: Qwen3.6 27B
60 tok/s+0%

Benchmarks

MMLU-ProNot evaluated
GPQA Diamond
57.6%
HLE
3.7%
LiveCodeBenchNot evaluated
SciCode
24.4%
TerminalBench Hard
8.3%
MATH-500Not evaluated
AIMENot evaluated
AIME 2025Not evaluated
IFBench
44.2%
Long Context Recall
30.7%
Tau2
20.8%
Market AverageTop Score

Open Source

Quick Compare

Similar Models

Compare all 7 models