vLLM
vLLM's own recipe page serves Mistral Medium 3.5 on nightly builds; architecture support had not yet shipped in a stable release.
SourceRetrieved 2026-08-12
Model reference
Mistral Medium 3.5
This reference covers the open merged flagship checkpoint and its restrictive license terms. It preserves each vendor date and currency without reconciling them.
Mistral Medium 3.5 is the vendor reference for mistralai/Mistral-Medium-3.5-128B, as of 2026-08-12. Model architecture: Dense 128B model with a 256k context window, handling instruction-following, reasoning, and coding; Context length: 256k, as of 2026-08-12. RunInfra has not measured this model; every figure below belongs to its named source, cited and dated.
Cited specifications
| Group | Fact | Source-cited display value | Source and date |
|---|---|---|---|
| Identity | Release identity | mistralai/Mistral-Medium-3.5-128B | SourceRetrieved 2026-08-12 |
| Identity | Vendor description | our first flagship merged model | SourceRetrieved 2026-08-12 |
| Identity | Announcement date | May 22, 2026 | SourceRetrieved 2026-08-12 |
| Identity | Documentation card date | April 28, 2026 | SourceRetrieved 2026-08-12 |
| Identity | Repository timing | The repository was created March 31, 2026. | SourceRetrieved 2026-08-12 |
| Architecture | Model architecture | Dense 128B model with a 256k context window, handling instruction-following, reasoning, and coding | SourceRetrieved 2026-08-12 |
| Architecture | Vision encoder | vision encoder trained from scratch | SourceRetrieved 2026-08-12 |
| Context | Context length | 256k | SourceRetrieved 2026-08-12 |
| Modalities | Inputs and output | Accepts both text and image input, with text output | SourceRetrieved 2026-08-12 |
| Modalities | Tool calling | Agentic: Best-in-class agentic capabilities with native function calling and JSON output | SourceRetrieved 2026-08-12 |
| Modalities | Reasoning control | Opt-in reasoning_effort with [THINK] blocks | SourceRetrieved 2026-08-12 |
| License | Weights license | Modified MIT License | SourceRetrieved 2026-08-12 |
| License | Revenue clause | You are not authorized to exercise any rights under this license if the global consolidated monthly revenue of your company (or that of your employer) exceeds $20 million (or its equivalent in another currency) for the preceding month. | SourceRetrieved 2026-08-12 |
| License | Derivative restriction | The revenue restriction extends to derivatives; a commercial license is available through Mistral. | SourceRetrieved 2026-08-12 |
| License | Vendor license framing | Released as open weights, under a modified MIT license | SourceRetrieved 2026-08-12 |
| Pricing | Documentation input price | EUR 1.25 per 1M tokens | SourceRetrieved 2026-08-12 |
| Pricing | Documentation output price | EUR 6.40 per 1M tokens | SourceRetrieved 2026-08-12 |
| Pricing | Announcement input price | $1.50 per 1M tokens | SourceRetrieved 2026-08-12 |
| Pricing | Announcement output price | $7.50 per 1M tokens | SourceRetrieved 2026-08-12 |
| Availability | Vendor hardware guidance | self-hosting possible on as few as four GPUs | SourceRetrieved 2026-08-12 |
| Availability | Merged flagship weights | mistralai/Mistral-Medium-3.5-128B | SourceRetrieved 2026-08-12 |
Vendor-claimed benchmarks
These results are vendor-claimed, not independently measured by RunInfra.
No vendor-claimed benchmark results for this exact build were present in the cited material as of 2026-08-12.
Serving support
Listed rows have a cited upstream support signal. They are not RunInfra measurements.
Serving concepts
What RunInfra measures when we measure it
RunInfra has not measured this model yet. When measurement is published, the record will state throughput, latency, memory, quality, serving conditions, and reproducible evidence.
Questions about this model reference
What is Mistral Medium 3.5?
As of 2026-08-12, release identity: mistralai/Mistral-Medium-3.5-128B.
Where are the Mistral Medium 3.5 weights?
As of 2026-08-12, vendor hardware guidance: self-hosting possible on as few as four GPUs.
Can I run Mistral Medium 3.5 myself?
As of 2026-08-12, vendor hardware guidance: self-hosting possible on as few as four GPUs.
RunInfraby RightNow© 2026 RunInfra. All rights reserved.
Ask AI about RunInfra
