Revision History
Small language model · 10 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 the Phi-4 benchmark comparison table and MATH score (56.1% was its GPQA score; MATH is 80.4%) and removed an unsupported 93.1% GSM8K figure; replaced a misattributed IBM definition quote; fixed 'Gemma Nano' (Google's on-device model is Gemini Nano), the Gemma 3n release date (June 2025, not September), Phi-2's '25x larger' claim, Phi-3-mini's token count (3.3T), Qwen2.5 small-model context length (32K), and an unsupported 100x Snapdragon NPU speedup (plus 24 minor fixes)
Version 10 · Nov 2, 2026, 08:48 AM
- References: renumbered 4 hidden references so they show and link
Version 9 · Oct 30, 2026, 09:46 AM
- Improve internal links (2026-10-13)
Version 8 · Oct 13, 2026, 10:08 AM
- Added 6 contextual internal links
Version 7 · Jul 23, 2026, 03:04 PM