Comprehensive side-by-side LLM comparison
Devstral Small 1.1 leads with 36.8% higher average benchmark score. Devstral Small 1.1 offers 239.6K more tokens in context window than DeepSeek-V2.5. Both models have similar pricing. DeepSeek-V2.5 is available on 3 providers. Overall, Devstral Small 1.1 is the stronger choice for coding tasks.
DeepSeek
DeepSeek-V2.5 was developed as an enhanced iteration of the DeepSeek-V2 architecture, designed to incorporate improvements in model quality and efficiency. Built to advance the DeepSeek foundation model series, it provides refined capabilities for general-purpose language understanding and generation tasks.
Mistral AI
Devstral Small was developed as a more efficient development-focused model, designed to bring software engineering assistance to resource-conscious deployments. Built to provide coding capabilities with reduced computational requirements, it enables developer tools in environments where efficiency is paramount.
1 year newer

DeepSeek-V2.5
DeepSeek
2024-05-08

Devstral Small 1.1
Mistral AI
2025-07-11
Cost per million tokens (USD)

DeepSeek-V2.5

Devstral Small 1.1
Context window and performance specifications
Average performance across 1 common benchmarks

DeepSeek-V2.5

Devstral Small 1.1
Performance comparison across key benchmark categories

DeepSeek-V2.5

Devstral Small 1.1
Available providers and their performance metrics

DeepSeek-V2.5
DeepInfra
DeepSeek
Hyperbolic

DeepSeek-V2.5

Devstral Small 1.1

DeepSeek-V2.5

Devstral Small 1.1

Devstral Small 1.1
Mistral AI