MiniMax-M2.5 vs North Mini Code 1.0

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.

North Mini Code 1.0 60

File safety 25, safetensors only, no .bin, .pt or .pth files and no Python in the repository. Provenance 15, the files sit in Cohere's verified CohereLabs organisation, which cohere.com links to per our other Cohere listings (10). Safety evaluation 0, no safety results on the card. Agent guidance 0, the card has no limitations or safety section, and nothing on prompt injection, instruction priority or what the model refuses. Licence risk 15, Apache 2.0 has no use policy, threshold or change clause. Security programme 5, Cohere's trust centre holds a responsible disclosure policy with a bug bounty, as recorded on our Cohere listings (5). 25 + 15 + 0 + 0 + 15 + 5 = 60.

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.

North Mini Code 1.0 85

Memory 25, the smallest build from Cohere is the w4a16 repository, 19.38 GB, and 19.38 GB × 1.2 = 23.3 GB, within 24 GB. Unsloth's GGUF isn't read for this line. Runtimes 20, Released vLLM 0.33.0, SGLang 0.5.23 and Transformers 5.22.2 have the cohere2_moe model, and llama.cpp b12242 has the cohere2moe architecture. Builds 15, Quantised builds from Cohere at two precisions, FP8 and 4-bit weights (w4a16), each in its own repository. Terms 15, Apache 2.0 in the metadata and on the card, an OSI licence with nothing added. Download 10, the Hub API reports gated false for the model and its builds on 10 November 2026, so the files download with no account and nothing to accept. Hosted 0, Hugging Face maps no inference provider to it. 25 + 20 + 15 + 15 + 10 + 0 = 85.

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.

North Mini Code 1.0 85

Chat template 20, chat_template.jinja in the repository (12,397 characters). Tool-calling format 25, the card documents how tools are declared (JSON schema through the chat template), how the call and its thinking go back into the history and how results are appended, and names the cohere_command4 tool-call parser for vLLM, which released vLLM has. Trained tool calling 20, the card says the model was specifically trained with tool-use capabilities for agentic coding, with reinforcement learning on verifiable tasks, and the template renders every call in a turn with thinking between them. Structured output 0, JSON schema is for describing tools. Reasoning control 10, the template has a reasoning_effort setting, and the card names the cohere_command4 reasoning parser, which returns thinking apart from the answer. Context 10, 256K tokens per the card, with 64K of output, and 500,000 positions in config.json. 20 + 25 + 20 + 0 + 10 + 10 = 85.