Mixtral-8x22B-Instruct-v0.1 is a large language model released by Mistral AI, released under the apache-2.0 license, with a claimed knowledge cutoff of 2024-01-31. Its weights, configuration, and tokenizer files have been cryptographically hashed and are queued for on-chain attestation by Weight Registry.
Description
Mistral's official instruct fine-tuned version of Mixtral 8x22B, a sparse Mixture-of-Experts model. It uses 39B active parameters out of 141B total, and its strengths include strong math, coding, and multilingual performance for a fully open (Apache 2.0) release of its size class.
Metadata
Hugging Face stats
30,950 downloads on Hugging Face · 758 likes · library: vllm
Weight Hashes
Weights file
SHA-256797fb7a6b1f07170c5538d5a2011557a109f6f85d4dcb849fd0f13b7087504ce
SHA3-256d3f2899af890607f37ffcb7a05eef4365e3be355019b3ab93c92b565dd5d18bd
BLAKE339e075dc00870c57d4b6ebb1a6a44ddd8dbdb81525c2902b571835dde5bc558c
Config file
SHA-2569c4a6138d84029ab666943613e3d5844d2ea8fd6149f44f77188c62e2915e0f5
SHA3-25683371e627739994ab4aef9b3b960751bc5b461341acd0029cb52e07f742a9cd9
BLAKE30f188c268ed0246643663c50dc196cd150cdd68b81dd3fbbfc95341c942529be
Tokenizer file
SHA-256e553af6fff7d7ad76e830608b218c5c0b0822998d5a1a96099a74cd3c1cb1a49
SHA3-256b35f60bdf1f8ed85174879e7e862ce60f953a406ecad9cbd62eee8a2969e34be
BLAKE3d635275f37a20dc55a8dc4996c7a0c51f92113b584e517bca3fcd03af021bbed
Verify it yourself
Download the file(s) from the source below, then hash them locally (sha256sum <file> on Linux/macOS, certutil -hashfile <file> SHA256 on Windows) and compare the output against the hash(es) shown above.