Verified65% confidenceFactExact time
MiniMax M3.1 Flash offers a 1M-token ultra-long context window.
3
Sources
65%
Confidence
Medium-term (~90 days)
Relevance
10/19/2026
First Seen
Valid until: 1/17/2027
Sources
Related Entities
Related Claims
UnverifiedMiniMax M3.1 Flash defaults to the max setting when the effort field is not passed in the request, and the model always performs thinking.74% similarUnverifiedEarly users report that MiniMax M3.1 Flash decodes at 90–110 tokens/second, but MiniMax has not yet published official speed data.74% similarUnverifiedMiniMax M3.1 Flash uses a MoE architecture with approximately 428B total parameters and approximately 23B activated per token.74% similarUnverifiedMiniMax M3.1 Flash's thinking tokens count toward the max_tokens limit, so reusing the 4K or 8K caps from the M2.x era may cause truncation with finish_reason: "length"; MiniMax recommends raising max_tokens to 131,072, with an upper limit of 524,288.73% similarUnverifiedCompared with MiniMax M2.7, M3.1 Flash's context window expands from 204,800 to 1,000,000, its input types expand from text-only to text, images, and video, and its thinking mode changes from always-on with no levels to five adjustable levels defaulting to max.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1067819API
curl https://kongchang.com/api/v1/knowledge/claims/1067819MCP
get_claim(id=1067819)