← EVENT LIBRARY / Foundation Models

MiniMax opens M3.1 Flash weights after its low-cost API launch

MiniMax M3.1 Flash is available through the API, M Plan and MiniMax Code at $0.15 per million input tokens and $0.60 per million output tokens. MiniMax has now also published a 252 GB, 48-shard checkpoint on Hugging Face under a custom license. Vendor speed and benchmark claims remain unverified independently.

Availability: MiniMax API, M Plan and MiniMax Code; public 252 GB checkpoint in 48 safetensor shards

What this means for you

For teams with high-volume coding or agent workloads, M3.1 Flash can now be tested both through the low-cost API and on controlled infrastructure. Self-hosting decisions must review the custom license and memory requirements, while quality claims still need workload-specific reproduction.

Evidence & related material

Subsequent changes

Correction · 2026-10-23

Follow-up verification confirms public M3.1 Flash weights, correcting the earlier unavailable status

Model weights were promised later and were recorded as unavailable at the cutoff. → A 252 GB, 48-shard checkpoint is public on Hugging Face under a custom license.

Technical details & scores
Availability MiniMax API, M Plan and MiniMax Code; public 252 GB checkpoint in 48 safetensor shards
Weights Published At HF initial commit 2026-10-23T03:37:20Z; evaluation metadata updated 2026-10-30T08:44:16Z
License custom MiniMax model license; review before production or redistribution
Pricing Usd Per Million Tokens
input 0.15 output 0.6 cached_input 0.015
Vendor Speed Claim 200+ TPS peak; 170+ TPS median
Vendor Benchmarks DeepSWE v1.1 73.7%; OSWorld V2 61.7; CyberGym 82%; evaluation details are vendor-reported
Time Basis The visible official page gives 2026-10-19 without a time; midnight metadata is not treated as a verified publication time.
Technical impact 4
Engineering value 4
Adoption signal 2
Maturity 4
Verification cost 2
Risk 3