What LLMs can an RTX 5090 run?

An RTX 5090 gives a model 32 GB of memory and 1,792 GB/s of bandwidth. Of the 40 open models tracked here, 18 fit at Q4_K_M with an 8,192-token context.

Models that fit on one RTX 5090

The most precise weights that still fit with 8K tokens of context and one request, the memory that takes, the longest context at that precision and the writing speed. Bigger models come first.

Model	Best precision	Memory	Longest context	Tokens/s
Qwen3.6 35B-A3B 36.0B, 3B active	GGUF Q6_K	30.9 GB	60K	157–291
Nemotron 3 Nano 30B-A3B 31.6B, 3.5B active	GGUF Q6_K	27.1 GB	256K (full)	144–266
Gemma 4 31B 31.3B	GGUF Q6_K	28.0 GB	101K	35–49
GLM-4.7 Flash 31.2B, 3B active	GGUF Q6_K	27.2 GB	92K	145–267
Xing 4.0 29B-A4B 31.2B, 4B active	GGUF Q6_K	27.1 GB	109K	120–218
Qwen3 30B-A3B 30.5B, 3.3B active	GGUF Q6_K	27.0 GB	40K (full)	125–226
Muse Glimmer 30B 29.8B	FP8 / INT8	31.2 GB	65K	31–44
Qwen3.6 27B 27.8B	GGUF Q8_0	31.3 GB	18K	31–44
Qwen3.8 27B 27.8B	GGUF Q8_0	31.3 GB	18K	31–44
Gemma 4 26B-A4B 25.8B, 4B active	GGUF Q8_0	28.9 GB	256K (full)	100–180
gpt-oss-20b 20.9B, 3.6B active	GGUF Q8_0	23.5 GB	128K (full)	111–200
Gemma 4 12B 12.0B	FP16 / BF16	25.4 GB	256K (full)	38–54
Qwen3.5 9B 9.7B	FP16 / BF16	20.6 GB	256K (full)	47–66
Qwen3 8B 8.2B	FP16 / BF16	18.5 GB	40K (full)	52–74
Llama 3.1 8B 8.0B	FP16 / BF16	18.1 GB	109K	53–75
Gemma 4 E4B 8.0B	FP16 / BF16	17.0 GB	128K (full)	56–80
Nemotron 3 Nano 4B 4.0B	FP16 / BF16	8.78 GB	256K (full)	103–154
MiniCPM5 2B 2.5B	FP16 / BF16	6.02 GB	128K (full)	144–222

Too big for one card

How many RTX 5090 cards these need at Q4_K_M in one tensor-parallel group; some need more than 8.

Llama 3.1 70B: 2 cards
Qwen3-Coder-Next: 2 cards
gpt-oss-120b: 4 cards
Nemotron 3 Super 120B-A12B: 4 cards
Qwen3.5 122B-A10B: 4 cards
Mistral Medium 3.5 128B: 4 cards
Qwen3.8 Flash Next: 4 cards
MiniMax M2.7: 8 cards
DeepSeek V4 Flash: 8 cards
DeepSeek V4 Flash 0731: 8 cards
MiMo V2.6 Flash: 8 cards
GLM-5.3 Flash: 8 cards
MiniMax M3: more than 8 cards
DeepSeek V3 / R1: more than 8 cards
DeepSeek V3.2: more than 8 cards
GLM-5.3: more than 8 cards
GLM-5.2: more than 8 cards
DeepSeek V4.1 Flash: more than 8 cards
MiMo V2.6 Pro: more than 8 cards
DeepSeek V4 Pro: more than 8 cards
Qwen3.8 2.4T-A95B: more than 8 cards
Kimi K3: more than 8 cards

How these numbers are worked out

Memory is the weights at each precision plus the KV cache for 8,192 tokens in FP16 and runtime overhead (0.5 GB plus 10%), computed from each model's files on Hugging Face with the LLM VRAM Calculator. Speed is estimated from memory bandwidth and the parameters read per token with the LLM Speed Calculator. Both tools take any other model from Hugging Face.

Other GPUs and Macs

Model numbers read from Hugging Face on 2026-09-26.