Mistral Large 4 vs Ornith-1.5-122B

Mistral Large 4 64

The card covers intended use, architecture, parameters, context and languages, and misses limitations and a cut-off (14). Training data is described by its mix in the technical report, with no sources, cut-off or filtering (5). The files changed on 2 November 2026, when the report was added (20). A Mistral member replied in three of nine discussions (10). It's a first release with older Large models in their own repositories and no dated note (5), and the files agree with each other and the card (10). 14 + 5 + 20 + 10 + 5 + 10 = 64.

Ornith-1.5-122B 36

Model card 11, it covers intended use, architecture, parameters and context but not limitations, languages or a knowledge cut-off. Training data 5, a description of the method with no data or cut-off. Last change 20, a folder upload to the MLX builds on 24 October 2026, 17 days ago. Responsiveness 0, Ornith answered none of the 5 questions and reports. Version policy 0, the MLX builds were uploaded a second time four days after the first with no note. File health 0, the card links a missing LICENSE, states hardware its own files don't fit and names a variant that doesn't exist, and the two chat templates differ. 11 + 5 + 20 + 0 + 0 + 0 = 36.

Mistral Large 4 50

Memory scores 0, since the smallest 4-bit build, Mistral's NVFP4, is 610 GB, about 732 GB with a fifth added, far above 256 GB. Runtimes score 10, released Transformers and llama.cpp have the architecture, while vLLM v0.33.0 and SGLang v0.5.23 still lack it, the llama.cpp build Ollama 0.40.14 pins predates it, and LM Studio has no entry. Builds score 15 for Mistral's FP8 and NVFP4 releases. Terms score 10, a modified MIT licence that is free commercially except for Model-as-a-Service businesses. Download scores 10 with no gate, and hosted too scores 5 because Mistral's API serves these weights. 0 + 10 + 15 + 10 + 10 + 5 = 50.

Ornith-1.5-122B 75

Memory 10, since Ornith's own MLX 4-bit build is 69.7 GB and needs about 83.6 GB with the fifth added, over 64 GB and inside the reference machine's 107.4 GB, as do its Q4_K_M GGUF, its NVFP4 build and Ollama's build. Runtimes 25, the most the line gives, from vLLM 0.33.0, SGLang 0.5.23, Transformers 5.22.2, llama.cpp's qwen35moe architecture, Ollama's library and LM Studio's catalogue for the qwen35moe architecture. Builds 15, Ornith publishes FP8, NVFP4, GGUF and MLX at several precisions. Terms 15, MIT as named in the card's metadata, with nothing added, read by its terms though the repository has no licence file. Download 10, nothing is gated. Hosted 0, Ornith runs no API and no inference provider serves it. 10 + 25 + 15 + 15 + 10 + 0 = 75.

Mistral Large 4 75

File safety scores 25, safetensors only and loaded by Transformers' own code with a Jinja template. Provenance scores 15, Mistral's verified organisation with a hash on every file. Safety evaluation scores 10, refusal and prompt-injection results Mistral ran itself, listed in its 2 November technical report as internal, and red-teaming by outside firms with nothing published. Agent guidance scores 10 for a statement of what the refusals cover and a claim of trained injection resistance, with only general safeguards. Licence risk scores 10, 5 off for the Model-as-a-Service limit. The security programme scores 5 for security.txt with no public advisories. 25 + 15 + 10 + 10 + 10 + 5 = 75.

Ornith-1.5-122B 50

File safety 25, safetensors and GGUF only and no Python in any of Ornith's repositories for this size, so the card's --trust-remote-code flag loads nothing from the repository. Provenance 15, the files come from the ornith-ai organisation Ornith's blog links to, with a SHA-256 for every weight file. Safety evaluation 0 and agent guidance 0, nothing published on either. Licence risk 10, MIT carries no policy, threshold or changeable terms, but the card's licence link is broken. Security programme 0, no security.txt, disclosure policy or security contact.

Mistral Large 4 95

The chat template is in the repository (20). The card documents how tools are declared and come back and names the parser (25), and the template renders several calls in one turn (20). Strict JSON Schema output is documented with how to ask for it (15). Reasoning effort is budgeted per request with the reasoning returned apart from the answer (10). Context scores 5, because the 1M window comes from YaRN scaling and the technical report gives 40K as the longest pretraining sequence. 20 + 25 + 20 + 15 + 10 + 5 = 95.

Ornith-1.5-122B 85

Chat template 20, a Jinja file in the repository. Tool-calling format 25, the card shows how tools go in and how calls come back and names the parser for vLLM and SGLang. Trained tool calling 20, Ornith trains it on agent tasks and the template renders several calls in one turn. Structured output 0, not documented or claimed. Reasoning control 10, Ornith's template reads enable_thinking per request and the card says the reasoning comes back in reasoning_content, though the card doesn't name the switch. Context 10, 262,144 tokens native.