ModelAI for Codev1.0

Codestral Mamba

by Mistral AI · open-source · Last verified 2026-03-17

Mistral AI's code-focused model built on the Mamba state-space architecture for linear-time inference scaling. Excels at long code sequences with theoretically unlimited context length and constant memory usage during generation.

https://mistral.ai/news/codestral-mamba ↗

C—Below Average

Adoption: CQuality: BFreshness: C+Citations: CEngagement: F

Specifications

License: Apache 2.0
Pricing: open-source
Capabilities: code-generation, code-completion, long-context-code, linear-time-inference, code-explanation
Integrations: huggingface, mistral-api, ollama, vllm
Use Cases: code-generation, large-codebase-analysis, code-completion, repository-understanding
API Available: Yes
Parameters: 7B
Context Window: 256K tokens
Modalities: text
Training Cutoff: Early 2024
Tags: code-generation, open-source, mamba-architecture, state-space-model, mistral
Added: 2026-03-17
Completeness: 92%

Index Score

41.2

Adoption

Quality

Freshness

Citations

Engagement

Explore the full AI ecosystem on Agents as a Service