MiniMax-M2.5
Official · BF16 · safetensors
Runtime certified: Transformers
| Precision / quantization | Format | Bits | Publisher trust | Deployment state |
|---|---|---|---|---|
| BF16 | safetensors | 16 | Official | Runtime certified |
MiniMax · General · MiniMax M2.5
MiniMax M2.5: Specs, GPU requirements, hosting & deployment
MiniMax M2.5 is a previous General release in the MiniMax family, certified from its immutable official source revision.
Total parameters230.0B
Active parameters10.0B
Context window196,608 tokens
Release statusPrevious · Previous
ArchitectureMiniMaxM2ForCausalLM
ModalitiesText
LicenseMiniMax Model License
Deployment stateDeployable with certified configuration
Commercial stateCommercial Use Allowed With Conditions
Release date2026-02-12
Architecture, strengths and use cases
MiniMax M2.5 is an official self-hostable MiniMax release in the MiniMax General track. It uses MiniMaxM2ForCausalLM, carries 230B / 10B MoE, supports a verified context window of 196,608 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.
MiniMax M2.5 uses MiniMaxM2ForCausalLM with 230B / 10B MoE. The stored architecture profile records 62 layers, hidden width 3072, 48 attention heads and a 196,608 token context ceiling where the official configuration publishes those fields.
Verified MiniMaxM2ForCausalLM architecture
196,608 token certified context
Official BF16, Unknown 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 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 memory441.26 GB
KV cache1.94 GB
Runtime overhead112.34 GB
Total with safety margin688.87 GB
Technically compatible reference sizing
No current FusionPointAI inventory passes every runtime, architecture, topology and memory gate for this exact checkpoint. Use the estimator for global compatible hardware.
License and self-hosting considerations
The full 230B 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.
MiniMax M2.5 is published under MiniMax Model 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
MiniMax M3, MiniMax M2.7, MiniMax M2.1
Release timeline
minimax/minimax/minimax-m2-5 @ dcbb3c5c80486ed208d15ef8fdc522a20eb11d69fb8018deef95d37452923d71
Frequently asked questions
How much VRAM does MiniMax M2.5 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 MiniMax M2.5?
Certified upstream runtime evidence is available for transformers.
Can MiniMax M2.5 be used commercially?
The stored commercial classification is Commercial Use Allowed With Conditions. Consult the official MiniMax Model License terms for the final obligations.
Official references
Last verified: 2026-08-26