Revision History
Mixtral · 13 revisions
Sizes are character counts of the article source. The signed number is the change from the previous revision.
Recent edit summaries
Detailed summaries recorded by editors. Generic maintenance summaries are omitted here; every recorded revision remains below. Dates describe the edit, not necessarily the event it covers.
- Correction: Corrected Mixtral's architecture (it does not use sliding window attention), the expert/shared parameter split, memory requirements, Mixtral 8x22B benchmark scores that did not match Mistral's announcement, a false claim that 8x22B served as mistral-large and that Le Chat's Mistral Small ran on Mixtral, an unsupported Le Chat Enterprise claim, wrong vLLM version and AWS date, wrong Qwen-72B/Falcon-180B MMLU and Qwen1.5-MoE license, a misattributed DeepSeek-V3 citation, an unsupported MLPerf vendor list, and other errors (plus 25 minor fixes)
Version 13 · Nov 2, 2026, 07:48 AM
- Improve internal links (2026-10-13)
Version 11 · Oct 13, 2026, 10:09 AM
- Added 6 contextual internal links
Version 10 · Jul 23, 2026, 03:14 PM
- Render formulas with LaTeX math notation
Version 9 · Jul 12, 2026, 12:37 AM