LLaVA 1.5 13B on AMD Radeon RX 9060 XT 8GB

Can the AMD Radeon RX 9060 XT 8GB run LLaVA 1.5 13B?

36 pts

Barely: most of the model spills into system RAM

Heavy offload. At Q2_K it needs about 9.7 GB of 8.0 GB usable, with an estimated speed of 20.9 tok/s.

Best quant: Q2_K
Generation speed: 20.9 tok/s
Prompt processing: 1,849 tok/s
First token (1K prompt): 0.6 s
Largest safe context: —

Download LLaVA 1.5 13B

Recommended for the AMD Radeon RX 9060 XT 8GB: Q2_K. Every other quantization has its own download link in the table below.

Quickest · Ollama

One command downloads a ready-to-run version and starts a chat.

ollama run llava:13b-v1.5-q2_K

Get Ollama at ollama.com/download (opens in a new tab) · model page (opens in a new tab)

Original weights · Hugging Face

Full-precision files from LLaVA, for fine-tuning, vLLM, Transformers.

huggingface.co/liuhaotian/llava-v1.5-13b (opens in a new tab)

No sign-in needed.

Links checked 1 week ago against Hugging Face, Ollama and GitHub.

How the D grade is scored

Every result gets up to 100 points from four parts. The points decide the letter grade.

Model qualityEditorial quality tier of the model, scaled by the quantization or precision used.	6.9 / 45
Memory fitFull marks when everything sits in GPU or unified memory; fewer when part spills to system RAM.	4.0 / 25
SpeedGeneration speed against a 40 tokens/s target for LLMs, or render time against a target for image and video.	10.8 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.	15.0 / 15
Total	36.8 / 100	36 ptsPoor (D) · 36 of 100 points: Model quality 6.9/45, Memory fit 4.0/25, Speed 10.8/15, Memory use 15.0/15

Grade scale: S 85+ · A 72+ · B 58+ · C 42+ · D 25+ · F under 25 points.

Every quantization on the AMD Radeon RX 9060 XT 8GB

Quant	Memory needed	Available	Speed	Fit	Grade · points	Download
`Q2_K`	9.7 GB	8.0 GB	20.9 tok/s	Heavy offload	36 ptsPoor (D) · 36 of 100 points: Model quality 6.9/45, Memory fit 4.0/25, Speed 10.8/15, Memory use 15.0/15
`Q3_K_M`	10.9 GB	8.0 GB	13.7 tok/s	Heavy offload	36 ptsPoor (D) · 36 of 100 points: Model quality 8.3/45, Memory fit 4.0/25, Speed 8.8/15, Memory use 15.0/15
`Q4_K_M`	12.5 GB	8.0 GB	9.3 tok/s	Heavy offload	35 ptsPoor (D) · 35 of 100 points: Model quality 9.4/45, Memory fit 4.0/25, Speed 7.2/15, Memory use 15.0/15
`Q5_K_M`	13.8 GB	8.0 GB	7.1 tok/s	Heavy offload	34 ptsPoor (D) · 34 of 100 points: Model quality 9.6/45, Memory fit 4.0/25, Speed 6.3/15, Memory use 15.0/15
`Q6_K`	15.3 GB	8.0 GB	5.7 tok/s	Heavy offload	34 ptsPoor (D) · 34 of 100 points: Model quality 9.8/45, Memory fit 4.0/25, Speed 5.7/15, Memory use 15.0/15
`Q8_0`	18.5 GB	8.0 GB	3.9 tok/s	Heavy offload	33 ptsPoor (D) · 33 of 100 points: Model quality 9.9/45, Memory fit 4.0/25, Speed 4.7/15, Memory use 15.0/15
`F16`	31.1 GB	8.0 GB	—	Won't run	2 ptsWon't run (F) · 2 of 100 points: Model quality 3.0/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15	Original weights (opens in a new tab)

Memory figures assume 4K of context. Longer conversations or documents grow the KV cache.

Other hardware that runs it

64 ptsGood (B) · 64 of 100 points: Model quality 9.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 NVIDIA GeForce RTX 3090 Q8_0 · 43.8 tok/s
64 ptsGood (B) · 64 of 100 points: Model quality 9.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 NVIDIA GeForce RTX 4090 Q8_0 · 47.2 tok/s
64 ptsGood (B) · 64 of 100 points: Model quality 9.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 AMD Radeon RX 7900 XTX Q6_K · 52.7 tok/s
64 ptsGood (B) · 64 of 100 points: Model quality 9.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 AMD Radeon RX 9070 XT Q4_K_M · 50.7 tok/s
64 ptsGood (B) · 64 of 100 points: Model quality 9.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 NVIDIA GeForce RTX 4070 Ti SUPER Q4_K_M · 61.8 tok/s
64 ptsGood (B) · 64 of 100 points: Model quality 9.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 NVIDIA GeForce RTX 4080 SUPER Q4_K_M · 67.7 tok/s

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.