GLM-4.6
GLM-4.6 is an open-weights AI model from Zhipu with an index score of 127 (GPT-4 = 100), rank 139 of 217.
127108–145Index score#139 of 217 models
$0.50/ $2.0Price per million tokensinput / output · via OpenRouter
205KContext windowtokens it reads at once
30 Sep 2025ReleasedReplaced by GLM-4.7
Where it sitson the frontier
o1 reached this score 10 months earlier.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Mathsevidence
Codingevidence
Agentsno evidence yet
Professional workevidence
Novel problemsno evidence yet
Long tasksno evidence yet
Third-party results only, at the model’s best standard setting. Hover a bar for who ran it.
What others sayrank on 8 leaderboards
The Model Indexthird-party tests, our method#139of 217 · 127
LMArena · Creative writingcreative writing#90of 360 · 1401LMArena · Textoverall chat quality, judged by people#105of 362 · 1424LMArena · Instruction followingfollowing detailed instructions#107of 362 · 1413LMArena · Long promptslong, complex requests#105of 340 · 1432LMArena · Mathsmaths questions in chat#110of 346 · 1418LMArena · Codingcoding questions in chat#122of 357 · 1458LMArena · WebDevbuilding web apps and front ends#107of 133 · 1342Best attop three among current models
Replaced by GLM-4.7; it no longer leads any category among current models.
Where to use itprices checked 18 Nov 2026
API providers 4 · per 1M tokens, in / out
Run it yourself
Official weightszai-org/GLM-4.6↗Host prices from OpenRouter; going direct can differ.
Facts and sources
Released30 Sep 2025
Announcementhuggingface.co ↗
Release entryGLM-4.6
Scored asGLM-4.6
Output speed–
Hugging Facezai-org/GLM-4.6 ↗