— Includes tokenizer fixes that were bugged in the initial version — quants done with an importance matrix for improved quantization loss — ggufs & imatrix generated from bf16 for «optimal» accuracy loss — Wide coverage of different gguf quant types from Q8 down to IQ1_S — Quantized with llama.cpp commit 5921b8f089d3b7bda86aac5a66825df6a6c10603 (master as of 2024-05-30) — Imatrix generated with this multi-purpose dataset by bartowski. Codestrall-22B-v0.1 is trained on a diverse dataset of 80+ programming languages, including the most popular ones, such as Python, Java, C, C++, JavaScript, and Bash (more details in the Blogpost). The model can be queried: — As instruct, for instance to answer any questions about a code snippet (write documentation, explain, factorize) or to generate code following specific indications — As Fill in the Middle (FIM), to predict the middle tokens between a prefix and a suffix (very useful for software development add-ons like in VS Code) It is recommended to use mistralai/Codestral-22B-v0.1 with mistral-inference. After installing mistralinference, a mistral-chat` CLI command should be available in your environment. Will generate an answer to…
Модальности:
Генерация текста
Области применения:
Генерация кода
Задача: Генерация текста
Автор: qwp4w3hyb
Теги: gguf, code, codestral, mistralai, mistral, imatrix, importance matrix, endpoints_compatible
Лайков: 3 | Загрузок: 351
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
