Mistral adds a third-party model, an Enterprise APIs tier, and new API tools
Mistral's API catalog grew: a third-party GLM 5.2 model, a quoted Enterprise APIs surcharge tier, a Priority inference speed, a free Leanstral coding agent, and a Data capture tool — while Vibe consumer plans held flat.
API catalog: Mistral open/premier models + Voxtral/OCR/embedding/agent tools; Standard/Batch inference speeds only; no named Enterprise API tier
API catalog adds GLM 5.2 (third-party), Enterprise APIs (+75% list price), Priority inference speed (~75% premium), Leanstral (free), and Data capture ($0.04/M tokens)
Proof of change
Mistral AI's pricing pages, as we captured them on two dates.
Showing the whole page as captured — scroll either panel, or open it at full size.
Show 6 other pages we compared
Showing the whole page as captured — scroll either panel, or open it at full size.
Showing the whole page as captured — scroll either panel, or open it at full size.
Showing the whole page as captured — scroll either panel, or open it at full size.
Showing the whole page as captured — scroll either panel, or open it at full size.
Showing the whole page as captured — scroll either panel, or open it at full size.
Showing the whole page as captured — scroll either panel, or open it at full size.
- These captures are 51 days apart, so anything that changed and reverted in between would not appear here.
- These prices come from our capture alone — they were not confirmed against an independent second source.
- 3 pages had no counterpart in the earlier capture and are not shown.
Both images are our own captures, taken on the dates shown. The pixels are unmodified — nothing is retouched, and where a region is highlighted the marker is drawn over the image, not into it. Open either capture to see it at full resolution.
Mistral widened its developer-API pricing page with several new surfaces. GLM 5.2, a third-party model from Zhipu/Z.ai, joins the “most powerful served models” lineup at $1.4 in / $4.4 out per million tokens ($0.14 cached input) — Mistral’s first publicly billed third-party model on la Plateforme. A new Enterprise APIs product is now named separately from Enterprise deployments: it is sales-gated and priced at +75% above list on select APIs in exchange for regional data-processing controls, system-level SLAs, and premium support.
The Standard/Batch inference-speed toggle gained a third option, Priority (~1.75x Standard rates, e.g. Mistral Small 4 Priority: $0.263 in / $1.05 out), for latency-sensitive workloads. Two new tools appeared: Leanstral, a free open-source Lean 4 code agent (Labs tag), and Data capture, which records and exposes API call data for debugging at $0.04 per million tokens. The document-extraction model also moved to version OCR 4.1 at unchanged pricing ($4 / $5 per 1,000 pages).
None of this touches the Vibe consumer/team subscription ladder (Free $0, Pro $14.99/mo, Team $24.99/user/mo, Enterprise custom) or the core flagship model rates (Mistral Medium 3.5, Mistral Small 4, Mistral Large 3), which are unchanged since the July 2026 price increase.
Live capture shows the API pricing page grew new surfaces since July: GLM 5.2 (a third-party Zhipu/Z.ai model, $1.4 in / $4.4 out per M tokens, $0.14 cached), a named 'Enterprise APIs' product (contact sales; +75% above list pricing on select APIs for regional data controls, system-level SLAs, premium support), a free open-source 'Leanstral' Lean 4 coding agent (Labs), and a new 'Data capture' tool ($0.04/M tokens). Mistral Moderation is now 'Mistral Moderation 2' and priced Free, down from $0.1/M tokens. A 'Priority' inference tier (~1.75x Standard rates) is now documented alongside Standard/Batch. OCR moved to version 4.1 at unchanged pricing ($4/$5 per 1,000 pages). Vibe consumer plans (Free/$14.99 Pro/$24.99 Team/Enterprise) are unchanged.