Model Comparison
Devstral Small 1.1 vs Llama 3.2 90B InstructWhich is better in 2026?
Comparing Devstral Small 1.1 and Llama 3.2 90B Instruct across benchmarks, pricing, and capabilities.
Verdict: Devstral Small 1.1 vs Llama 3.2 90B Instruct — which is better?
Devstral Small 1.1 (by Mistral AI) and Llama 3.2 90B Instruct (by Meta) are two of the AI models people compare most. Here is how they stack up on benchmarks, price and capabilities, and which one to pick in 2026.
On price, Devstral Small 1.1 is roughly 2.4x cheaper per token on a blended 3:1 input/output basis, which adds up quickly at production volume.
Choose Devstral Small 1.1 if…
- cost matters — it's about 2.4x cheaper per token
- you want the most recent training data — it shipped Jul 2025
Choose Llama 3.2 90B Instruct if…
- you want predictable pricing at $0.35/M input and $0.40/M output
Performance Benchmarks
Comparative analysis across standard metrics
Devstral Small 1.1 and Llama 3.2 90B Instructdon't have any common benchmark datasets to compare. They may have been evaluated on different testing suites.
Arena Performance
Human preference votes
Pricing Analysis
Price comparison per million tokens
For input processing, Devstral Small 1.1 ($0.10/1M tokens) is 3.5x cheaper than Llama 3.2 90B Instruct ($0.35/1M tokens).
For output processing, Devstral Small 1.1 ($0.30/1M tokens) is 1.3x cheaper than Llama 3.2 90B Instruct ($0.40/1M tokens).
In conclusion, Llama 3.2 90B Instruct is more expensive than Devstral Small 1.1.*
* Using a 3:1 ratio of input to output tokens
Model Size
Parameter count comparison
Llama 3.2 90B Instruct has 66.0B more parameters than Devstral Small 1.1, making it 275.0% larger.
Context Window
Maximum input and output token capacity
Both models have the same input context window of 128,000 tokens. Both models can generate responses up to 128,000 tokens.
Input Capabilities
Supported data types and modalities
Llama 3.2 90B Instruct supports multimodal inputs, whereas Devstral Small 1.1 does not.
Llama 3.2 90B Instruct can handle both text and other forms of data like images, making it suitable for multimodal applications.
Devstral Small 1.1
Llama 3.2 90B Instruct
License
Usage and distribution terms
Devstral Small 1.1 is licensed under Apache 2.0, while Llama 3.2 90B Instruct uses Llama 3.2.
License differences may affect how you can use these models in commercial or open-source projects.
Apache 2.0
Open weights
Llama 3.2
Open weights
Release Timeline
When each model was launched
Devstral Small 1.1 was released on 2025-07-11, while Llama 3.2 90B Instruct was released on 2024-09-25.
Devstral Small 1.1 is 10 months newer than Llama 3.2 90B Instruct.
Jul 11, 2025
1.0 years ago
9mo newerSep 25, 2024
1.8 years ago
Knowledge Cutoff
When training data ends
Neither model specifies a knowledge cutoff date.
Unable to compare the recency of their training data.
Provider Availability
Devstral Small 1.1 is available from Mistral AI. Llama 3.2 90B Instruct is available from DeepInfra, Bedrock, Fireworks, Together, Hyperbolic.
Devstral Small 1.1
Llama 3.2 90B Instruct
Outputs Comparison
Key Takeaways
Devstral Small 1.1
View detailsMistral AI
Detailed Comparison
Interactive Arena
Judge for yourself.
Run your own prompts against Devstral Small 1.1 and Llama 3.2 90B Instruct side-by-side, then vote on the output you prefer.
| Feature | | |
|---|
FAQ
Common questions about Devstral Small 1.1 vs Llama 3.2 90B Instruct.