AI Coding Models
Every model we track, grouped by the company that builds it. 25 of 26 are scored on the LLM Coding Leaderboard; the rest are waiting to be tested.
OpenAI
-
GPT-5.6-Luna
#3 -
GPT-5.6-Sol
#1
Anthropic
Tencent
-
Tencent Hy3
#5 -
Tencent Hy4 Preview
Not testedThis is a preview model by Tencent, announced on August 28, 2026. I have not run the full LLM coding benchmark with it, because, after just 1 prompt out of 20,...
Moonshot AI
-
Kimi K3
#9 -
Kimi K2.7 Code
#13
xAI
Z.ai
-
GLM-5.3
#11 -
GLM-5.3-Flash
#28 -
GLM-5.2
#25
DeepSeek
-
Gemini-3.6-Flash
#23 -
Gemini-3.7-Flash
#17 -
Gemini-3.1-Pro
#26
Alibaba
-
Qwen 3.8 27B
#18 -
Qwen 3.8 Max
#22
Meta
-
Muse Spark 1.2
#24
Stealth
-
Ox Alpha
#27This model was a stealth codename for the model GLM-5.3-Flash, released on August 26, 2026.
MiniMax
-
Minimax M3
#30
Cursor
-
Composer 2.5
#31
Placements come from the LLM Coding Leaderboard, where every tested model is compared side by side on the same projects. Models tested at several thinking levels show their best placement here.