llm11

Codestral 2508 with Jev

Codestral 2508 is a model from Mistral, released 2025-08-01. It costs $0.300 per million input tokens and $0.900 per million output tokens, reads up to 256K tokens of context, and writes up to 204,800 tokens in one reply. Through llm11, set model to mistralai/codestral-2508.

Input price
$0.300 / 1M
Output price
$0.900 / 1M
Context window
256K tokens
Max output
205K tokens
Cached input
$0.030 / 1M

Read from the provider's prompt cache

Released
2025-08-01
Accepts
text, file

Read from the live catalogue and refreshed daily. At blended list price, 63% of the 332 priced models we can call cost more.

Using Codestral 2508 with Jev

Jev is TypeSafe AI's System One model. It reads each request first and decides which model in your pool answers, with a calibrated confidence on that call. Codestral 2508 does the answering when Jev sends it work, or when you name it yourself. Codestral 2508 falls in the llm11-fast band, and the labs and limits it meets make it eligible for that pack. Packs keep a spread of eight models per band, and it is not one of the eight today, so it is only picked when you name it in a custom pool.

Jev is new, so we have not published results for Codestral 2508 with and without it, and this page does not invent any. Every response comes with a receipt that names the model that answered and what it cost against your most expensive model, so you can measure the pairing on your own traffic.

About JevThe System One router

What a request costs

List price arithmetic, the same sum a receipt does. llm11 adds nothing per request; the only fee is on buying credits.

RequestTokens in / outOneA thousand
Short chat turn500 / 200$0.00033$0.330
Question over retrieved documents4,000 / 500$0.00165$1.65
Long document summary30,000 / 1,000$0.00990$9.90

Cheaper models to try

The catalogue has no benchmark score for Codestral 2508, so this is a price ordering. It does not claim these answer as well.

Cheaper alternatives to Codestral 2508
ModelInput / 1MOutput / 1MContext
Qwen Qwen3.5-35B-A3B$0.163$1.30262K
DeepSeek DeepSeek V3.1$0.250$0.950164K
Qwen Qwen3.6 Flash$0.188$1.131M
Qwen Qwen3 Next 80B A3B Thinking$0.150$1.20262K

Call it

Same OpenAI request shape. Naming the model pins it, so nothing is routed, and the answer is still checked.

python

from openai import OpenAI

client = OpenAI(base_url="https://www.llm11.com/v1", api_key="llm11_live_...")

res = client.chat.completions.create(
    model="mistralai/codestral-2508",
    messages=[{"role": "user", "content": "Hello"}],
)

More from Mistral

Other Mistral models
ModelInput / 1MOutput / 1MContext
Mistral Medium 3.5$1.50$7.50262K
Mistral Small 4$0.150$0.600262K
Devstral 2 2512$0.400$2.00262K
Ministral 3 14B 2512$0.200$0.200262K
Ministral 3 3B 2512$0.100$0.100131K
Ministral 3 8B 2512$0.150$0.150262K

All 19 Mistral models

Questions

What is Codestral 2508 with Jev?
Jev is TypeSafe AI's System One model. It reads each request first and decides which model in your pool answers, with a calibrated confidence on that call. Codestral 2508 does the answering when Jev sends it work, or when you name it yourself. Codestral 2508 falls in the llm11-fast band, and the labs and limits it meets make it eligible for that pack. Packs keep a spread of eight models per band, and it is not one of the eight today, so it is only picked when you name it in a custom pool. Routing decisions are made by Jev, TypeSafe AI's System One model.
Is Codestral 2508 a System One model?
No. Codestral 2508 answers requests. The System One model is Jev, which decides which model answers each one, so the two do different jobs and llm11 uses both.
How much does Codestral 2508 cost?
$0.300 per million input tokens and $0.900 per million output tokens. A 4,000 token prompt with a 500 token answer costs about $0.00165, so a thousand of them is about $1.65. llm11 passes provider list price through and charges 5% when you buy credits.
What is the Codestral 2508 context window?
256,000 tokens, with up to 204,800 tokens in a single reply.
Does Codestral 2508 support tool calling and structured output?
The catalogue lists tool calling and structured output, but it does not list reasoning controls.
Can I use Codestral 2508 through llm11?
Yes. Set model to mistralai/codestral-2508 on the OpenAI-compatible endpoint and the request goes to it directly, with verification still running.
Does Jev route to Codestral 2508?
Routing decisions are made by Jev, TypeSafe AI's System One model. Codestral 2508 falls in the llm11-fast band, and the labs and limits it meets make it eligible for that pack. Packs keep a spread of eight models per band, and it is not one of the eight today, so it is only picked when you name it in a custom pool.
What is a cheaper alternative to Codestral 2508?
Nearest in price and below it: Qwen Qwen3.5-35B-A3B at $0.163 in and $1.30 out, DeepSeek DeepSeek V3.1 at $0.250 in and $0.950 out and Qwen Qwen3.6 Flash at $0.188 in and $1.13 out. The catalogue has no benchmark score for Codestral 2508, so this is a price ordering. It does not claim these answer as well.