Mistral's Mamba Codestral B is an open-source code model built on the Mamba architecture. It is designed for tasks such as code generation, function calling, and structured output.
Input
Output
Context
256K
Max Output
256
Parameters
7.3B
Input Modalities
Output Modalities
Some values come from web research and may be incorrect. Report incorrect data
None the catalog can show from stated facts or measured scores.
Download 29.1 GB · 5 GGUF builds
| Precision | Weights | One request at 33K tokens | Smallest GPU that fits |
|---|---|---|---|
| FP16 / BF16 | 13.6 GB | 17.7 GB | 1x NVIDIA GeForce RTX 3090 (24 GB) |
| INT8 | 6.79 GB | 10.6 GB | 1x NVIDIA T4 (16 GB) |
| INT4 | 3.39 GB | 7.03 GB | 1x NVIDIA T4 (16 GB) |
Computed by Inferbase from the 7,285,403,648 parameters in the creator's Hugging Face repository; one request adds its KV cache, activations and framework overhead, with the layer layout estimated from the parameter count.
Public resources
Specification sources
As of Nov 10, 2026.
Answered from the facts on this page, with their sources.
Mamba Codestral 7B V0.1 has a context window of 256,000 tokens and a maximum output of 256 tokens (source: Web research). A provider may serve less; Inferbase routes on what each provider serves.
Yes. Mamba Codestral 7B V0.1 is an open-weight model under the Apache 2.0 license. You can download and self-host it within the license terms.
Mamba Codestral 7B V0.1 has 7,285,403,648 parameters (7.3B), as stated in the creator's Hugging Face repository.
Strengths: Open weights under a permissive license (apache-2.0). Each is computed from the model's stated facts and benchmark results.
Mamba Codestral 7B V0.1 accepts Text and produces Text. The Capabilities card lists the API features it supports, such as function calling and structured output.
No benchmark source has published a result for Mamba Codestral 7B V0.1 yet, so it is shown as unmeasured, not estimated.
Not on Inferbase Managed today. Add your own provider key or endpoint that serves Mamba Codestral 7B V0.1, and Inferbase runs it on your account with the same routing and logs.
The weights take 13.6 GB at FP16 / BF16, 6.8 GB at INT8, 3.4 GB at INT4, computed from the 7,285,403,648 parameters its creator publishes. Each request adds its KV cache and the framework's overhead. The hardware section shows one request and the smallest GPU that fits; the GPU planner sizes it for your context length and traffic.
Start on the free plan with 5,000 requests a month, no card required.