Mistral-Medium-3.5-128B
mistralai/Mistral-Medium-3.5-128B
Official · FP8 · safetensors
Runtime certified: Transformers vLLM SGLang
| Precision / quantization | Format | Bits | Publisher trust | Deployment state |
|---|---|---|---|---|
| FP8 | safetensors | 8 | Official | Runtime certified |
Mistral · Generalist Medium · Mistral Medium 3.5
Mistral Medium 3.5 128B: Specs, GPU requirements, hosting & deployment
Mistral Medium 3.5 128B is a current Generalist Medium release in the Mistral family, certified from its immutable official source revision.
Total parameters128.0B
Active parametersDense model
Context window262,144 tokens
Release statusCurrent · Current
ArchitectureMistral3ForConditionalGeneration
ModalitiesText
LicenseModified MIT License
Deployment stateDeployable with certified configuration
Commercial stateCommercial Use Allowed With Conditions
Release date2026-03-31
Architecture, strengths and use cases
Mistral Medium 3.5 128B is an official self-hostable Mistral AI release in the Mistral Generalist Medium track. It uses Mistral3ForConditionalGeneration, carries 128B, supports a verified context window of 262,144 tokens, and exposes only source-certified checkpoints and runtime combinations. Its release, license and deployment facts are tied to the official repository revision rather than inferred from the family name.
Mistral Medium 3.5 128B uses Mistral3ForConditionalGeneration with 128B. The stored architecture profile records 88 layers, hidden width 12288, 96 attention heads and a 262,144 token context ceiling where the official configuration publishes those fields.
Verified Mistral3ForConditionalGeneration architecture
262,144 token certified context
Official FP8 checkpoint coverage
Private LLM inference
Agentic workflows
Long-context document processing
Certified checkpoints and quantizations
Only exact, source-certified variants attached to this release are shown.
Recommended hardware
Use only the listed transformers, vllm, sglang runtime configuration with the exact checkpoint precision. The recommendation engine checks weight memory, KV cache, runtime overhead, safety headroom, GPU architecture and topology before ranking rentable and reference hardware.
Estimated weight memory122.79 GB
KV cache1.38 GB
Runtime overhead40.47 GB
Total with safety margin204.15 GB
Recommended best value
2× NVIDIA H200 SXM 141GB
282 GB total VRAM · nvswitch
77.85 GB headroom · Limited capacity
USD 4,690 / monthLowest-cost viable
2× NVIDIA H200 SXM 141GB
282 GB total VRAM · nvswitch
77.85 GB headroom · Limited capacity
USD 4,690 / monthHighest performance
2× NVIDIA H200 SXM 141GB
282 GB total VRAM · nvswitch
77.85 GB headroom · Limited capacity
USD 4,690 / monthLicense and self-hosting considerations
The full 128B resident weight set must fit across the selected GPUs. Context and concurrency add KV-cache memory; MoE active parameters do not replace resident-weight memory.
Mistral Medium 3.5 128B is published under Modified MIT License. The release-scoped official license link and verification date are retained with the catalogue record.
Commercial Use Allowed With Conditions. Commercial use is conditional. Review the linked official terms before deployment.
Alternatives and sibling variants
Mistral-Medium-3.5-128B, Mistral-Medium-3.5-128B-NVFP4, Mistral-Medium-3.5-128B-EAGLE
Release timeline
mistral-ai/mistral/mistral-medium-3-5-128b @ bd6dc6b9d18a72d3a09e68492438a99b4c0bc8ced688686458f2e9e2116d5a87
Frequently asked questions
How much VRAM does Mistral Medium 3.5 128B require?
The exact estimate depends on checkpoint precision, context and concurrency; the page recommendation uses the certified memory engine and compatible runtime configurations.
Which runtimes support Mistral Medium 3.5 128B?
Certified upstream runtime evidence is available for transformers, vllm, sglang.
Can Mistral Medium 3.5 128B be used commercially?
The stored commercial classification is Commercial Use Allowed With Conditions. Consult the official Modified MIT License terms for the final obligations.
Official references
Last verified: 2026-08-26