Released 2026-06-01 · general agent
Deployment evidence
MiniMax · approximately 428B total · approximately 23B active · 1M tokens
Best-fit scenarios
Multimodal agent, coding and long-context workloads with explicit commercial-license review.
Not recommended
- Commercial launch before license notice and attribution review
- Stable-production serving without pinned framework versions
Multimodal agent workflows
Vendor-statedOfficial materials position M3 for multimodal perception and agent tasks.
Capability sourceCoding and tool orchestration
Public evidenceThe official release reports coding and agent evaluations.
Capability sourceCommercial embedded assistant
ChinaAPI inferencePotentially suitable after the revenue, notice and attribution clauses are cleared.
Official launch / minimum
Evidence DCandidate recipe under review
A 4x RTX PRO 6000 NVFP4 recipe is under review upstream; V1 does not promote an unmerged recipe to validated minimum.
200-person team
Evidence EBenchmark required
A quantized multi-GPU worker is plausible, but the 200-seat recommendation needs measured TTFT, throughput and context mix.
Commercial API
Evidence BFramework support maturing
Aggregated and disaggregated vLLM recipes are still landing; pinning a nightly build may be required.
Commercial-use check
MiniMax Community License
MaaS / hosted service: Commercial API and hosted use are Commercial Use. Above USD 20M yearly revenue obtain prior written authorization; otherwise send the required one-time notice.
Attribution: Prominently display Built with MiniMax M3 for commercial use.
Prohibited uses: The license includes specified unlawful, military and harmful-use restrictions.
Known limitations and open questions
- Stable vLLM release support was not complete at the V1 cutoff
- Do not equate a pending recipe with successful ChinaAPI reproduction