💡

Complete Your AI Tool Stack

Mistral Medium 3.5 users also rely on these tools to enhance their workflow:

💰 Affiliate disclosure: We may earn a commission if you sign up through these links at no extra cost to you.

Part of 781+ curated AI tools on AISO
Mistral Medium 3.5 logo

Mistral Medium 3.5

Mistral's 128B merged flagship — open weights, coding+reasoning+instructions

0
freemiumDR 86API: $1.5/M input tokens, $7.5/M output tokens. Available free on Le Chat. Open weights on Hugging Face (modified MIT license). Self-hostable on 4 GPUs.View full pricing →

Visit Mistral Medium 3.5

https://mistral.ai/news/vibe-remote-agents-mistral-medium-3-5/

About Mistral Medium 3.5

Mistral's first flagship merged model, released May 22, 2026. A dense 128B model with a 256k context window that handles instruction-following, reasoning, and coding in a single set of weights. Available as open weights (modified MIT license) and powers Mistral Vibe remote coding agents and Le Chat's new Work mode. SWE-Bench Verified: 77.6%. API: $1.5/M input, $7.5/M output.

Key Features

128B dense model (merged: instruction-following + reasoning + coding)
256k token context window
77.6% on SWE-Bench Verified (beats Devstral 2 and Qwen3.5 397B A17B)
91.4 on τ³-Telecom (strong agentic capabilities)
Configurable reasoning effort per request
Vision encoder trained from scratch — handles variable image sizes and aspect ratios
Open weights under modified MIT license (self-hostable on 4 GPUs)
Powers Mistral Vibe remote coding agents and Le Chat Work mode
Async cloud coding sessions with GitHub, Linear, Jira, Sentry integrations

Mistral Medium 3.5 Pros & Cons

Pros

  • +Open weights under modified MIT license — genuinely self-hostable on 4 GPUs, unlike most frontier models
  • +Strong coding benchmark: 77.6% on SWE-Bench Verified, outperforming Devstral 2 and Qwen3.5 397B A17B
  • +Merged model architecture means instruction-following, reasoning, and coding in one set of weights — no routing overhead
  • +Configurable reasoning effort per request — tune latency vs. quality per call
  • +256k context window handles long documents and extended agentic sessions
  • +Competitive API pricing: $1.5/$7.5 per million tokens (vs Claude Opus 4.8 at $5/$25)
  • +Powers Mistral Vibe remote agents — async cloud coding with Slack, GitHub, Linear, Jira integrations

⚠️ Cons

  • 128B dense model requires significant GPU resources for self-hosting at scale (4 GPUs minimum for inference)
  • Open weights use a modified MIT license — not standard MIT, review commercial terms for enterprise use
  • Newer and less battle-tested than GPT-4o or Claude Sonnet in production agentic pipelines
  • Vision capabilities are present but Mistral has not published detailed vision benchmarks for Medium 3.5
  • Vibe remote agents and Work mode are on Pro/Team/Enterprise plans — not available on free Le Chat

Who Is Mistral Medium 3.5 Best For?

👤Developers who want a frontier-class coding model they can self-host
👤Teams building agentic workflows with long-horizon coding tasks
👤Organizations in regulated industries needing on-premise deployment
👤Budget-conscious API users who need strong reasoning at lower token cost than Claude/GPT-4o

Tags

mistralllmapiopen-weightscodingreasoningagenticself-hostedfrontier model
🏷️

Is this your tool?

Claim your listing to get a Featured badge, edit your description, and stand out from competitors. All plans include a permanent dofollow backlink to your site.

Claim Now →

📬 Get the best new AI tools delivered weekly

One concise email with fresh launches, trending picks, and featured standouts.

Alternatives to Mistral Medium 3.5

View all Mistral Medium 3.5 alternatives →