GLM-5
GLM-5 is an open-weights AI model from Zhipu with an index score of 142 (GPT-4 = 100), rank 88 of 217.
142126–157Index score#88 of 217 models
$1.0/ $3.2Price per million tokensinput / output · via OpenRouter
205KContext windowtokens it reads at once
11 Feb 2026ReleasedReplaced by GLM-5.1
Where it sitson the frontier
o3 reached this score 10 months earlier.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Mathsevidence
Codingevidence
Agentsno evidence yet
Professional workevidence
Novel problemsevidence
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 9 leaderboards
The Model Indexthird-party tests, our method#88of 217 · 142
LMArena · Creative writingcreative writing#39of 360 · 1447LMArena · Textoverall chat quality, judged by people#56of 362 · 1457LMArena · Long promptslong, complex requests#55of 340 · 1469LMArena · Instruction followingfollowing detailed instructions#61of 362 · 1447LMArena · Codingcoding questions in chat#74of 357 · 1498LMArena · Mathsmaths questions in chat#77of 346 · 1443Epoch Capabilities Indexoverall capability from many benchmarks#87of 231 · 146.5LMArena · WebDevbuilding web apps and front ends#80of 133 · 1435Best attop three among current models
Where to use itprices checked 18 Nov 2026
API providers 8 · per 1M tokens, in / out
Run it yourself
Official weightszai-org/GLM-5↗Host prices from OpenRouter; going direct can differ.
Facts and sources
Released11 Feb 2026
Announcementhuggingface.co ↗
Release entryGLM-5
Scored asGLM-5
Output speed–
Hugging Facezai-org/GLM-5 ↗