SunsetDiff

Replacement for Codestral Mamba 7B (0.1): what changes when you switch

The official replacement is codestral-2405. Codestral Mamba 7B (0.1) (open-codestral-mamba) shut down on 2025-06-06 per Mistral's deprecation page (checked 2026-10-17). The published context window changes from 256,000 tokens on open-codestral-mamba to 32,000 tokens on codestral-2405.

Codestral Mamba 7B (0.1) model page · Mistral

Official replacement. Full diff page

Codestral Mamba 7B (0.1) to Codestral (24.05)

Some values not published Checked 2026-10-17.

Price per 1M tokens

Price comparison. Example monthly cost for 10M input tokens and 2M output tokens. Labeled example, not a bill.
FieldOldNewChange
Input Not published $0.3 Not published
Output Not published $0.9 Not published
Cached input Not published $0.03 Not published
Example monthly cost Not published $4.8 Example only

Context and max output

Token limits
LimitOldNewChange
Context window 256,000 32,000 -87.5% (lower)
Max output Not published Not published Not published

Capabilities that differ

No differing capability values are published for both models.

API changes

No parameter change is listed on the official pages we checked. The sample below only swaps the model id.

Code

python before
from openai import OpenAI
client = OpenAI()
response = client.chat.completions.create(
    model="open-codestral-mamba",
    messages=[{"role": "user", "content": "Hello"}],
)
python after
from openai import OpenAI
client = OpenAI()
response = client.chat.completions.create(
    model="codestral-2405",
    messages=[{"role": "user", "content": "Hello"}],
)
typescript before
import OpenAI from "openai";
const client = new OpenAI();
const response = await client.chat.completions.create({
  model: "open-codestral-mamba",
  messages: [{ role: "user", content: "Hello" }],
});
typescript after
import OpenAI from "openai";
const client = new OpenAI();
const response = await client.chat.completions.create({
  model: "codestral-2405",
  messages: [{ role: "user", content: "Hello" }],
});
curl before
curl https://api.openai.com/v1/chat/completions \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"open-codestral-mamba","messages":[{"role":"user","content":"Hello"}]}'
curl after
curl https://api.openai.com/v1/chat/completions \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"codestral-2405","messages":[{"role":"user","content":"Hello"}]}'

Checklist

  1. Read the shutdown date for open-codestral-mamba on the linked official page.
  2. Point your requests at codestral-2405.
  3. Re-check price, context, and parameters on the provider pages linked here.
  4. Ship the change before the published shutdown date.

Related models

Questions

What is the replacement for Codestral Mamba 7B (0.1)?

The official replacement is codestral-2405. Codestral Mamba 7B (0.1) (open-codestral-mamba) shut down on 2025-06-06 per Mistral's deprecation page (checked 2026-10-17). The published context window changes from 256,000 tokens on open-codestral-mamba to 32,000 tokens on codestral-2405.

Is codestral-2405 cheaper than open-codestral-mamba?

A full price comparison is not published. One or both prices are missing from the official pages we checked.

Does codestral-2405 have a bigger context window than open-codestral-mamba?

No. The context window changes from 256,000 to 32,000 tokens.

Sources

Deprecation alerts

We store your email for deprecation alerts. We do not send email.

Deprecation alert signup
Providers (optional)