Skip to content

What can the DGX Station GB300 run?

What fits on the DGX Station GB300
Text models: 83Visual models: 0Strongest first
Memory
252 GB
Bandwidth
7,100 GB/s
Maker
NVIDIA

Open the calculator on this machineDGX Station GB300 in our inventory

Text models that run on it

Text models that run on the DGX Station GB300, best first
ModelBest fitContextSpeedStrongest at
Kimi K31.6-bit1M20 tok/sChat, Coding, Agents, Docs
MiMo V2.6 Pro1.6-bit832K168 tok/sChat, Coding, Agents, Docs
DeepSeek V4.1 Flash1.6-bit1M322 tok/sChat, Coding, Agents, Docs
GLM-5.31.6-bit960K158 tok/sChat, Coding, Agents, Docs
GLM-5.3 Flash5-bit1M188 tok/sChat, Coding, Agents, Docs
Qwen3.8 2.4T A95B1.6-bit256K22 tok/sChat, Coding, Agents
Qwen3.8 Flash-Next8-bit256K239 tok/sChat, Coding, Agents
Muse Spark Open 816B A42B1.6-bit1M183 tok/sChat, Docs
MiMo V2.6 FlashFP4 + FP81M198 tok/sCoding, Agents
MiniMax M33-bit352K171 tok/sDocs
DeepSeek V4 FlashFP4 + FP81M157 tok/sDocs
DeepSeek V4 Pro2-bit1M29 tok/sChat, Coding, Agents, Docs
Kimi K2.61.6-bit256K200 tok/sChat, Coding, Agents, Docs
Hy36-bit40K121 tok/sChat, Coding, Agents, Docs
Nemotron 3 Ultra 550B A55B2-bit256K93 tok/sChat, Coding, Docs
Qwen3.5 397B A17B4-bit256K175 tok/sAgents, Docs
MiniMax M2.7FP896K174 tok/sCoding, Agents
DeepSeek V3.22-bit52K155 tok/sChat, Coding, Agents, Docs
Mistral Medium 3.5 128BFP8256K34 tok/sChat, Coding, Agents, Docs
Qwen3.5 122B A10B8-bit256K198 tok/sChat, Coding, Agents, Docs
Nemotron 3 Super 120B A12B8-bit256K127 tok/sChat, Coding, Agents, Docs
Nemotron 3.5 Bolt 74B A9B8-bit256K138 tok/sChat, Coding, Agents, Docs
Mistral Small 4 119BFP8256K287 tok/sChat, Docs
Qwen3 Coder Next 80B8-bit256K283 tok/sCoding, Agents
Llama 3.3 70B8-bit128K58 tok/sChat, Coding, Agents, Docs
Inkling1.6-bit4K177 tok/sText and images
DeepSeek R12-bit160K150 tok/sNot rated
DeepSeek V32-bit160K150 tok/sNot rated
GLM 51.6-bit198K155 tok/sNot rated
GLM 4.5 Air8-bit128K168 tok/sNot rated
Inkling Small6-bit4K227 tok/sText and images
MiniMax M2.5FP896K165 tok/sNot rated
GLM 5.11.6-bit198K155 tok/sNot rated
Laguna M.18-bit52K109 tok/sNot rated
Laguna S 2.18-bit1M213 tok/sNot rated
Qwen3 VL 235B A22B Instruct8-bit28K97 tok/sText and images
GLM 5.21.6-bit960K156 tok/sNot rated
Qwen3 235B A22B8-bit32K97 tok/sNot rated
Ornith 1.5 122B8-bit256K177 tok/sNot rated
Kimi K2 Instruct1.6-bit128K186 tok/sNot rated
Ornith 1.5 397B4-bit256K151 tok/sNot rated
Ornith 1.0 397B4-bit256K172 tok/sNot rated
Step 3.5 Flash8-bit104K212 tok/sNot rated
DeepSeek V3 03242-bit160K150 tok/sNot rated
Qwen2.5 72B Instruct8-bit32K57 tok/sNot rated
Qwen3 Next 80B A3B Instruct8-bit256K283 tok/sNot rated
DeepSeek R1 05282-bit160K150 tok/sNot rated
GLM 4.54-bit96K101 tok/sNot rated
Llama 4 Scout 17B 16E Instruct8-bit672K148 tok/sText and images
MiniMax M2FP896K165 tok/sNot rated
MiMo V2.56-bit52K174 tok/sNot rated
Qwen2.5 VL 72B Instruct8-bit125K56 tok/sText and images
Solar Open 100B8-bit128K183 tok/sNot rated
GLM 5.3 Flash BF165-bit1M187 tok/sText and images
Qwen3 VL 235B A22B Thinking8-bit28K97 tok/sText and images
Qwen3 235B A22B Instruct 25078-bit32K97 tok/sNot rated
NVIDIA Nemotron 3.5 Super VL 120B A12B BF168-bit1M127 tok/sText and images
Hy3 Preview6-bit32K121 tok/sNot rated
Qwen3 Coder 480B A35B Instruct3-bit80K140 tok/sCoding
GLM 4.74-bit96K101 tok/sNot rated

Best fit is the highest quality, up to 8-bit, that leaves the use case’s context. Ranked by our use-case scores, then by Hugging Face downloads. Speeds are rough, plus or minus 25%. The top 60 of 83 are shown.

Questions

What is the best LLM for the DGX Station GB300?

Of the open models we track, the strongest that run well on the DGX Station GB300 are Kimi K3, MiMo V2.6 Pro, DeepSeek V4.1 Flash. 83 text models fit it in all.

Calculated from published specs, not benchmarked. All machines