MiniMax-M2.5 vs Step 3.7 Flash

MiniMax-M2.5 30

File safety earns 15, safetensors only with no pickle files, less 10 because every documented loading route passes trust_remote_code for the repository's Python. Provenance earns the full 15, files in the organisation MiniMax links to with a SHA-256 for each shard. Safety evaluation and agent guidance earn nothing, as MiniMax publishes neither. Licence risk earns nothing (read 21 October 2026). The Prohibited Uses Policy goes beyond the law, MiniMax can end the licence for any breach and replaced its terms three weeks after release, and that change on 6 March 2026 came with no dated note while the card still links to a deleted file. The security programme earns nothing, with no security.txt, disclosure policy, SECURITY.md or advisories, and only a general contact address on the card.

Step 3.7 Flash 45

File safety 15, safetensors and GGUF only, no .bin, .pt or .pth files (25). Provenance 15, the files sit in the stepfun-ai organisation, which the card's StepFun platform and model page belong to (10). Safety evaluation 0, no safety results on the card. Agent guidance 0, the card has no limitations section and gives no guidance on prompt injection, instruction priority, refusals or safeguards. Licence risk 15, Apache 2.0 has no use policy, threshold or change clause. Security programme 0, no way to report a flaw on the card, and our StepFun API listing found no security.txt, bug bounty or disclosure policy. 15 + 15 + 0 + 0 + 15 + 0 = 45.

MiniMax-M2.5 75

The chat template earns 20 for chat_template.jinja in the repository. The tool-calling format earns 25, as MiniMax's own guide documents how tools go in and how calls come back, gives a parsing function, and its deployment guides name the vLLM and SGLang parsers. Trained tool calling earns 20, with several invoke blocks in one call. Structured output earns nothing, since JSON output isn't documented or claimed. Reasoning control earns nothing, because thinking is always on with no documented way to turn it off or budget it. Context earns 10 for 196,608 tokens with no position scaling.

Step 3.7 Flash 85

Chat template 20, chat_template.jinja in the repository (5,723 characters). Tool-calling format 25, the card names the parser to use in two runtimes, step3p5 in vLLM (with auto tool choice) and in SGLang, and both ship Step parsers in a release. Trained tool calling 20, the card presents tool orchestration as a main use, with multi-turn results, and the template renders every call in a turn. Structured output 0, Nothing on the card about JSON or schema-constrained output for the weights. Reasoning control 10, the template has a reasoning_effort setting and keeps reasoning_content, and the card names the step3p5 reasoning parser for vLLM and SGLang, which returns the reasoning apart from the answer. Context 10, 256K tokens per the card, and 262,144 positions in config.json. 20 + 25 + 20 + 0 + 10 + 10 = 85.

MiniMax-M2.5 65

Memory earns 5. MiniMax only publishes its 230 GB FP8 release, and the smallest build of 4-bit or better from a runtime's own library is LM Studio's MLX 4-bit at 128.7 GB, about 154 GB with the fifth added, so it needs more than our 107.4 GB machine and fits within 256 GB. Runtimes earn the full 25, with model code in released vLLM v0.33.0, SGLang v0.5.23, llama.cpp and Transformers, an LM Studio catalogue entry, and Ollama 0.40.14, whose pinned llama.cpp b11963 has the minimax-m2 architecture, though Ollama's library lists no build. Builds earn 10 for one publisher precision, the FP8 release. Terms earn 10 for a custom licence that allows free commercial use under a use policy. Download earns 10, with no gate or account. Hosted earns 5, since Novita and Featherless AI serve it through Hugging Face, while MiniMax's own API is closed to new users and retires it on 18 November 2026. 5+25+10+10+10+5 = 65.

Step 3.7 Flash 70

Memory 5, the smallest build at 4 bits is StepFun's IQ4_XS GGUF, 104.99 GB with the 3.97 GB projector, and 108.96 GB × 1.2 = 130.8 GB, over what the reference machine can give a model and within 256 GB. The IQ3 and Q3 builds are below 4 bits and aren't read. Runtimes 20, Released vLLM 0.33.0, SGLang 0.5.23 and Transformers 5.22.2 have the step3p7 model, and llama.cpp b12242 has the architecture, with StepFun's own GGUF. Four at 5 each, and the card documents all four. Builds 15, StepFun publishes GGUF at five quantised precisions, an FP8 build and an NVFP4 build. Terms 15, Apache 2.0 in the card's metadata and its licence section, an OSI licence with nothing added. Download 10, the Hub API reports gated false for all four repositories on 10 November 2026, so the files download with no account and nothing to accept. Hosted 5, StepFun's own API serves it, per the card and our StepFun API listing, and Featherless and DeepInfra are live for it on Hugging Face. 5 + 20 + 15 + 15 + 10 + 5 = 70.