Model Comparison

Comprehensive side-by-side analysis of model capabilities and performance

DeepSeek-V2.5

DeepSeek

DeepSeek-V2.5 is a language model developed by DeepSeek. It achieves strong performance with an average score of 71.1% across 15 benchmarks. It excels particularly in GSM8k (95.1%), MT-Bench (90.2%), HumanEval (89.0%). The model is available through 3 API providers. Released in 2024, it represents DeepSeek's latest advancement in AI technology.

Gemma 3 1B

Google

Gemma 3 1B is a language model developed by Google. The model shows competitive results across 18 benchmarks. It excels particularly in IFEval (80.2%), GSM8k (62.8%), Natural2Code (56.0%). It's licensed for commercial use, making it suitable for enterprise applications. Released in 2025, it represents Google's latest advancement in AI technology.

DeepSeek-V2.5

DeepSeek

2024-05-08

Gemma 3 1B

Google

2025-03-12

10 months newer

Performance Metrics

Context window and performance specifications

DeepSeek-V2.5

Larger context

Max Context:16.4K

Parameters:236.0B

Gemma 3 1B

Max Context:-

Parameters:1.0B

Average performance across 3 common benchmarks

DeepSeek-V2.5

+35.5%

Average Score:86.3%

Gemma 3 1B

Average Score:50.8%

Performance comparison across key benchmark categories

DeepSeek-V2.5

math

+42.7%

84.9%

general

+50.6%

68.4%

code

+23.1%

66.1%

Gemma 3 1B

math

42.2%

general

17.8%

code

43.0%

Benchmark Scores - Detailed View

Side-by-side comparison of all benchmark scores

Provider Availability & Performance

Available providers and their performance metrics

DeepSeek-V2.5

3 providers

Hyperbolic

Throughput: 100 tok/s

Latency: 0.5ms

DeepInfra

Throughput: 63 tok/s

Latency: 0.5ms

DeepSeek

Throughput: 100 tok/s

Latency: 0.5ms

Gemma 3 1B

0 providers

DeepSeek-V2.5

+35.5%

Avg Score:86.3%

Providers:3

Gemma 3 1B

Avg Score:50.8%

Providers:0