DeepSeek-Coder-V2-Instruct
DeepSeek- Arena score
- 1191
- Among open models
- #113of 148
- Among all on Arena
- #328
- Votes
- 15,147
Measured on DeepSeek-Coder-V2, the served version, not a local build. LMArena, CC BY 4.0, published Nov 11, 2026.
Score by category
CategoryScoreOpen rankArena rankVotes
- Coding1251#100#2972.7K
- Math1241#97#2871.9K
- Software & IT1239#105#3054.3K
- Applied math1233#99#2941.7K
- Long prompts1219#107#3111.8K
- Hard prompts, English1217#106#3102.7K
- Hard prompts1207#106#3124.4K
- Chinese1201#106#3111.7K
- English1194#116#3388.1K
- Russian1188#111#3201.8K
- French1185#85#261207
- Non-English1182#110#3217K
- Expert1181#104#301769
- Instructions1180#111#3245.6K
- Multi-turn1176#111#3262.7K
- Writing & literature1170#114#3343.9K
- German1164#88#274516
- Legal & government1163#117#334888
- Science1163#116#3382.6K
- Business & finance1155#113#3301.8K
- Spanish1153#93#274245
- Medicine & health1133#115#331847
- Japanese1127#73#248345
- Media & sports1122#119#3432.5K
- Creative writing1120#121#3512.3K
- Korean1104#80#258399
- Excluding ties1070#112#3279.9K
About as strong, smaller
Compare
granite-4.2-3bIBM Granite12923.7B
gemma-3-4b-itGoogle12914.3B

gemma-3n-E4B-itGoogle13067.8B

aya-expanse-8bCohere Labs11858.0B
Discussion
Built on it
Ways to run it
- GGUF · llama.cpp, Ollama, LM Studiolegraphista/DeepSeek-Coder-V2-Instruct-IMat-GGUF7 builds
- MLX · Apple silicon: mlx-lm, LM StudioGGorman/DeepSeek-Coder-V2-Instruct-Q4-mlx4 builds
Most downloaded
- DeepSeek-Coder-V2-Instruct-IMat-GGUFlegraphista · Quantization2.6K downloads
- DeepSeek-Coder-V2-Instruct-GGUFbartowski · Quantization1.1K downloads
- DeepSeek-Coder-V2-Instruct-GGUFsecond-state · Quantization1.1K downloads
- DeepSeek-Coder-V2-Instruct-GGUFgaianet · Quantization815 downloads
- DeepSeek-Coder-V2-Instruct-iMat-GGUFqwp4w3hyb · Quantization381 downloads
BenchClub runs
Not runnable yet
Runs open once a DeepSeek-Coder-V2-Instruct build is registered.
More from DeepSeek
See more
DeepSeek-V3.29.6M685B

DeepSeek-V4-Flash-07312M304B

DeepSeek-V4.1-Flash1.6M763B

DeepSeek-V4-Flash1.2M291B
Other 80–300B models
See more
MiniMax-M2.5MiniMax1M229B
Inkling-SmallThinking Machines1M266B
Hy3Tencent987.2K299B

NVIDIA-Nemotron-3-Super-120B-A12B-BF16NVIDIA874.7K124B