This model is deprecated. We only continue performance benchmarking for the default 10k input token workload. Results for other workloads are historical and no longer updated.

Google has launched a newer model, Gemma 4 E4B. We suggest considering it instead.

For more information, see comparison of Gemma 4 E4B to other models and API provider benchmarks for Gemma 4 E4B.

Open weights model

Released March 2025

Gemma 3 4B Instruct API Provider Benchmarking & Analysis

Model Comparison

This analysis is intended to support you in choosing the best API provider of Gemma 3 4B Instruct for your use-case.

Fastest

#1
AmazonAmazon
188.8 t/s
#2
Parasail (FP8)Parasail (FP8)
79.2 t/s
#3
DeepInfraDeepInfra
20.5 t/s

Output speed

Total 4 providers

Lowest Latency

#1
AmazonAmazon
1.02 s
#2
DeepInfraDeepInfra
1.19 s
#3
Parasail (FP8)Parasail (FP8)
1.29 s

Time to first token

Total 4 providers

Lowest Price

#1
Google (AI Studio)Google (AI Studio)
$0.00
#2
AmazonAmazon
$0.04
#3
Parasail (FP8)Parasail (FP8)
$0.05
#4
DeepInfraDeepInfra
$0.06

Blended price (per 1M tokens)

Total 4 providers

Gemma 3 4B is available through 4 API providers, each offering different performance characteristics and pricing. Below is a comparison of the key metrics across providers.

  • For output speed, the top providers are Amazon (188.8 t/s), Parasail (FP8) (79.2 t/s), and DeepInfra (20.5 t/s).
  • For latency, Amazon (1.02s), DeepInfra (1.19s), and Parasail (FP8) (1.29s) offer the lowest time to first token.
  • For pricing, Google (AI Studio) (0.00), Amazon (0.04), and Parasail (FP8) (0.05) offer the lowest blended prices per 1M tokens.
  • Amazon offers the best performance with both the highest speed and lowest latency. For cost optimization, Google (AI Studio) provides the most competitive pricing.

Highlights

Output tokens per second · Higher is better
Seconds · Lower is better
USD per 1M tokens (blended) · Lower is better

Pricing

Pricing: Cache Hit, Input, and Output

Price (USD per M Tokens) · Lower is better · 10,000 input tokens

Pricing: Blended Price

Blended at 7:2:1 (cache-input-output) · USD per 1M tokens (blended) · Lower is better

Output Speed vs. Price

Blended at 7:2:1 (cache-input-output) · Output speed: output tokens per second · USD per 1M tokens (blended)
Most attractive quadrant

Speed

Measured by Output Speed (tokens per second)

Output Speed: Gemma 3 4B Instruct

Output speed: output tokens per second · 10,000 input tokens

Latency vs. Output Speed

Latency: seconds to first token received · Output speed: output tokens per second · 10,000 input tokens
Most attractive quadrant

Latency

Measured by Time (seconds) to First Token

Time to First Token

Seconds to first token received · Lower is better · 10,000 input tokens

End-to-End Response Time

Seconds to output 500 tokens, calculated based on time to first token, 'thinking' time for reasoning models, and output speed

End-to-End Response Time

Seconds to output 500 tokens, including reasoning model 'thinking' time · Lower is better · 10,000 input tokens

Key Comparison Metrics & API Features

Amazon Bedrock logoAmazon Bedrock
128k
Open
--
189
1.02
3.67
--
Google logoGoogle
128k
Open
--
--
--
--
--
Parasail logoParasail
131k
Open
--
77
1.29
7.78
--
DeepInfra logoDeepInfra
131k
Open
--
20
1.23
26.48
--

Frequently Asked Questions

Common questions about Gemma 3 4B Instruct providers