Mistral-7B-v0.1 is a large language model released by Mistral AI, released under the Apache-2.0 license, with a claimed knowledge cutoff of 2023-09-27. Its weights, configuration, and tokenizer files have been cryptographically hashed and are queued for on-chain attestation by Weight Registry.
Description
A 7.3B-parameter dense transformer using grouped-query and sliding-window attention for faster inference at long context, matching or exceeding LLaMA 2 13B's benchmarks despite being roughly half the size -- released fully permissively under Apache 2.0 with no usage restrictions, becoming a default base model for open-source fine-tuning.
Metadata
Hugging Face stats
430,912 downloads on Hugging Face · 4,184 likes · pipeline: text-generation · library: transformers
Weight Hashes
Weights file
SHA-256e4d1ddfac91bebd4bd571a90aa48c834a70bef8c6a5c0b4b4533f3654d174be4
SHA3-25615be2dd62836cda5a7dff1545b55d938bdd4be731d5aacaf8c2a972a2ce7f3cf
BLAKE38a0808966716f9cd0a45adee321ebc8a43c17c6036fe504b7c9a99d0f8b238c2
Config file
SHA-256cf25cdf4719f181d1d1d371973285d9afe9afde0d0c6a6fd48de857555ce1e0d
SHA3-2565f4ce1355846a776ce13eee193f3a552eb2c61140399d7c605ef05eb78619968
BLAKE3c3b53a834918949302eae3b689506b4cd2791b14f0522d0b0299464f074f58ca
Tokenizer file
SHA-25611c08db21487c885d8c792180f0be237f6a261b89a46f128a6a80a3aa4bd1720
SHA3-2560e81dce2e2032ac2dfc26adf7a1fe0f6a685569971191e5b2968a81c1b3c10f5
BLAKE30d8c8f93f99c57d58d54ea6c81562579dd5a04de4bd1f8ff900250ec69b34a0e
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.
Anomalies Flagged
- other weight-like files present but not hashed: ['pytorch_model-00001-of-00002.bin', 'pytorch_model-00002-of-00002.bin']