AI Briefing#minimax
The latest AI and developer news about #minimax, with the original source and a short summary.
The latest AI and developer news about #minimax, with the original source and a short summary.
MiniMax has previewed its M3 model, which introduces a new sparse attention mechanism to achieve 15.6x faster long-context response speed.
MiniMax plans to release M3, a multimodal model using Deepseek-style attention, as an open weight model.
MiniMax AI has teased the release of its next-generation model, MiniMax-M3, through official channels.
In an RTX 5090 Llama.cpp experiment, MiniMax-M2.7 processed at 22 tps even at 100k context but only generated at 3~4 tps.
In Terminal-Bench 2.0, MiniMax M2.7 scored 41.3%, lower than M2.5.
Opus 4.7 has clearly improved, while Kimi K2.6-Code-Preview remains a wait-and-see.
A new clause has been added to the MiniMax-M2.7 model license permitting free use for personal and non-profit research purposes.
MiniMax M2.5 has been added to the free and student usage tiers of GitHub Copilot.
The perplexity NaN cause in some MiniMax M2.7 GGUF was investigated and fixed.
Running MiniMax M2.7 AWQ on 2 GX10s to complete a local agentic coding stack.
Frontier open models require sustained funding, so they will likely end up moving toward a consortium.
MiniMax has unveiled MMX-CLI, its official multimodal CLI for agents.
GLM-5 and MiniMax M2.7 delivered closed-model-level performance on core agentic tasks.
MiniMax M2.7 has been added to Vercel AI Gateway in two versions, standard and high-speed.
MiniMax M2.5, a model that supports the entire development process, is now available in AI Gateway.
Qwen3-TTS now supports natural language voice design and 3-second voice cloning.