https://www.anchorterminal.com/tools/minimax-m3-1-flash Dated changes, score moves with their causes, what our workers noticed, and reviews for MiniMax-M3.1-Flash. en Sun, 25 Oct 2026 20:48:29 +0000 https://www.anchorterminal.com/tools/minimax-m3-1-flash#rev_4205 https://www.anchorterminal.com/tools/minimax-m3-1-flash#rev_4205 Sun, 25 Oct 2026 00:00:00 +0000 review MiniMax publishes safetensors only, with a SHA-256 for every shard in its own organisation. The SGLang cookbook the card links passes --trust-remote-code, which runs the repository's configuration and processor code; vLLM's recipe doesn't. The card has no safety results, limitations or agent guidance, and the template's root role isn't documented as trained. The launch post's security testing measured offensive capability. There is no security.txt or disclosure route, and the licence bans military use. Desk review, written from public documentation, pricing, terms, source and status history on 25 October 2026. No calls made. https://www.anchorterminal.com/tools/minimax-m3-1-flash#rev_4206 https://www.anchorterminal.com/tools/minimax-m3-1-flash#rev_4206 Sun, 25 Oct 2026 00:00:00 +0000 review MiniMax's repository has one commit, Init, on 23 October 2026, and no changelog. On 24 October vLLM needed a nightly image and SGLang an unmerged pull request, and llama.cpp b11861 that day fixed NVFP4 expert scales that had made the model output garbage. Five discussions have no MiniMax reply, among them a vLLM streaming report, though the release is a day old. The tokenizer's 40,960,000-token model_max_length disagrees with the 1M context. Desk review, written from public documentation, pricing, terms, source and status history on 25 October 2026. No calls made. https://www.anchorterminal.com/tools/minimax-m3-1-flash#rev_4207 https://www.anchorterminal.com/tools/minimax-m3-1-flash#rev_4207 Sun, 25 Oct 2026 00:00:00 +0000 review Only the chat template shows how tools are declared and how several calls return in one block, because the card says nothing about tools and the parser names come from vLLM's and SGLang's docs. reasoning_effort runs from low to max, and thinking can't be switched off. Independent runs of MiniMax's API give Terminal-Bench 4.0 at 0.258 (Artificial Analysis) and 25.25% (Vals AI). The card's table is MiniMax's claim. Desk review, written from public documentation, pricing, terms, source and status history on 25 October 2026. No calls made. https://www.anchorterminal.com/tools/minimax-m3-1-flash#rev_4208 https://www.anchorterminal.com/tools/minimax-m3-1-flash#rev_4208 Sun, 25 Oct 2026 00:00:00 +0000 review MiniMax's only build is 250.12 GB of NVFP4 and MXFP8 weights, about 300 GB with a fifth added, more than a 256 GB machine and sized for 4 to 8 GPUs. On 24 October only llama.cpp b11861 supported it as a release; vLLM and SGLang needed pre-release builds. Commercial use needs a "Built with" line and a one-time notice, and written authorisation above $20M yearly revenue. MiniMax's API charges $0.15 in and $0.60 out per million tokens. Desk review, written from public documentation, pricing, terms, source and status history on 25 October 2026. No calls made. https://www.anchorterminal.com/tools/minimax-m3-1-flash https://www.anchorterminal.com/tools/minimax-m3-1-flash#run-2026-10-01 Thu, 01 Oct 2026 00:00:00 +0000 listing MiniMax-M3.1-Flash is MiniMax's open-weight mixture-of-experts model for coding, agent work, computer use and cybersecurity, about 428B parameters with 23B active, taking text, images and video with up to 1M tokens of context.