Skip to main content

codestral-2508 vs gpt-oss-120b

Pricing, Performance & Features Comparison

Authormistral

Context Length256K

Reasoning

-

Providers1

ReleasedAug 2025

Knowledge Cutoff–

License–

Codestral 25.08: Mistral’s coding model for high-frequency completion and code-generation workloads.

Input$0.3

Output$0.9

Latency (p50)3.9s

Output Limit–

Function Calling

JSON Mode

-

InputText

OutputText

in$0.3out$0.9––

Latency (24h)

Success Rate (24h)

Authoropenai

Context Length131K

Reasoning

-

Providers0

ReleasedAug 2025

Knowledge Cutoff–

License–

OpenAI most powerful open weight model, which fits into a single H100 GPU.

Input–

Output–

Latency (p50)–

Output Limit131K

Function Calling

JSON Mode

-

InputText

OutputText

–