From 06cca7093ca5a5cb094e97d00a4cc51f25c89d91 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Patrik=20Bj=C3=B6rklund?= Date: Thu, 29 Feb 2024 07:39:29 +0100 Subject: [PATCH] [PUBLISHER] Merge #69 --- content/posts/mistral-large.md | 29 +++++++++++++++++++++++++++++ 1 file changed, 29 insertions(+) create mode 100644 content/posts/mistral-large.md diff --git a/content/posts/mistral-large.md b/content/posts/mistral-large.md new file mode 100644 index 0000000..1867bf4 --- /dev/null +++ b/content/posts/mistral-large.md @@ -0,0 +1,29 @@ +--- +title: Mistral Large +authors: + - Patrik Björklund +share: true +date: 2024-02-29 07:35:00 +0200 +summary: Mistral releases a new model. Mistral large. Does not seem to beat GPT-4 +tags: + - AI +series: + - AI +--- + +Mistral has launched its latest AI model, Mistral Large, boasting enhanced reasoning and multilingual capabilities. + +My key question: Is Mistral Large more adept at writing semi-complicated Python scripts than its competitor, GPT-4? + +I did a straightforward, non-scientific test (have two chats with it) to determine the models' proficiency in generating Python code with specific instructions. + +Verdict: +> GPT-4 maintains its edge in accurately writing Python scripts that adhere to the given instructions. + +Mistral's introduction of both Mistral Large and Mistral Small models marks a significant step in their commitment to providing versatile AI solutions and provides healthy competition. + +Mistral Large's performance in various benchmarks highlights its potential in multiple domains, even as it falls short in direct comparison with GPT-4 for most things. + +The company has also announced partnerships and plans for future enhancements, including improvements in function calling and JSON formatting, promising a comprehensive toolset for developers and organizations. + +Good to see - but not a true shakeup. \ No newline at end of file