invokeable()
← comparisons / deepseek-v4-pro-0813-vs-mistral-medium-2604

DeepSeek V4 Pro 0813 vs Mistral Medium 3.5

Side-by-side from the Invokeable directory — provider-attributed pricing, canonical IDs, draft data flagged. Open as a shareable card →

01Head to head

DeepSeek V4 Pro 0813Mistral Medium 3.5
CreatorDeepSeekMistral AI
Released2026-08-122026-04-29
Context1M262K
Max output384K262K
Input $/M——
Output $/M——
Knowledge cutoff——
Modalitiestext → texttext + image → text
Capabilitiestool-use, extended-thinking, structured-outputtool-use, extended-thinking, structured-output
ProvidersDeepSeek APIMistral AI (La Plateforme)

02Benchmarks · vendor-reported unless verified

DeepSeek V4 Pro 0813Mistral Medium 3.5
SWE-bench Verified (resolved)—77.6
τ³-Telecom (accuracy)—91.4