Check my GPU

LLMs ·

InternVL3 78B

Parameters
78B
Architecture
Dense
Max context
32K tokens
License
qwen
Released
2025-04

How to run it

Quantized files · GGUF

Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.

huggingface.co/unsloth/InternVL3-78B-GGUF (opens in a new tab)

ollama run hf.co/unsloth/InternVL3-78B-GGUF

Original weights · Hugging Face

Full-precision files from Shanghai AI Lab, for fine-tuning, vLLM, Transformers.

huggingface.co/OpenGVLab/InternVL3-78B (opens in a new tab)

No sign-in needed.

Links checked 7 days ago against Hugging Face, Ollama and GitHub.

Memory needed

Weights plus KV cache and runtime overhead. Leave a little extra free for the operating system and other apps.

Quantization Quality Weights Total at 8K context Total at 32K context Download
Q2_K Low 31.4 GB 34.6 GB 42.7 GB GGUF (opens in a new tab)
Q3_K_M Low 38.4 GB 41.7 GB 49.8 GB GGUF (opens in a new tab)
Q4_K_M Good 47.8 GB 51.1 GB 59.2 GB GGUF (opens in a new tab)
Q5_K_M High 55.7 GB 59.0 GB 67.0 GB GGUF · folder (opens in a new tab)
Q6_K High 64.3 GB 67.6 GB 75.6 GB GGUF · folder (opens in a new tab)
Q8_0 Very high 83.1 GB 86.4 GB 94.5 GB GGUF (UD-Q8_K_XL) · folder (opens in a new tab)
F16 Full 157 GB 160 GB 168 GB GGUF · folder (opens in a new tab) Original weights (opens in a new tab)

Best hardware for InternVL3 78B

The least expensive GPU in our catalog that runs it without offloading is the AMD Radeon PRO W7800 48GB (launch price $2,799) at Q3_K_M.

Grade · points Hardware Quant Memory Speed Fit
82 ptsGreat (A) · 82 of 100 points: Model quality 31.6/45, Memory fit 25.0/25, Speed 12.6/15, Memory use 13.4/15 NVIDIA RTX PRO 6000 Blackwell Q4_K_M 55.9 GB / 96.0 GB 28.1 tok/s Runs well
78 ptsGreat (A) · 78 of 100 points: Model quality 33.0/45, Memory fit 25.0/25, Speed 5.4/15, Memory use 15.0/15 Apple M4 Max (128 GB) Q6_K 75.3 GB / 96.0 GB 5.1 tok/s Runs well
77 ptsGreat (A) · 77 of 100 points: Model quality 33.0/45, Memory fit 25.0/25, Speed 6.5/15, Memory use 12.7/15 Apple M2 Ultra (192 GB) Q6_K 79.1 GB / 144 GB 7.5 tok/s Runs well
77 ptsGreat (A) · 77 of 100 points: Model quality 33.0/45, Memory fit 25.0/25, Speed 4.1/15, Memory use 15.0/15 NVIDIA GB10 (128 GB) Q6_K 76.8 GB / 115 GB 3.0 tok/s Runs well
76 ptsGreat (A) · 76 of 100 points: Model quality 33.0/45, Memory fit 25.0/25, Speed 3.7/15, Memory use 15.0/15 AMD Ryzen AI Max+ 395 (128 GB) Q6_K 75.3 GB / 96.0 GB 2.4 tok/s Runs well
74 ptsGreat (A) · 74 of 100 points: Model quality 33.0/45, Memory fit 25.0/25, Speed 6.6/15, Memory use 10.0/15 Apple M3 Ultra (256 GB) Q6_K 82.9 GB / 192 GB 7.7 tok/s Runs well
74 ptsGreat (A) · 74 of 100 points: Model quality 28.0/45, Memory fit 24.0/25, Speed 7.3/15, Memory use 15.0/15 Apple M4 Max (64 GB) Q3_K_M 45.5 GB / 48.0 GB 9.5 tok/s Tight fit
73 ptsGreat (A) · 73 of 100 points: Model quality 28.0/45, Memory fit 24.0/25, Speed 6.3/15, Memory use 15.0/15 Apple M1 Max (64 GB) Q3_K_M 45.5 GB / 48.0 GB 7.0 tok/s Tight fit
70 ptsGood (B) · 70 of 100 points: Model quality 33.0/45, Memory fit 25.0/25, Speed 6.6/15, Memory use 5.9/15 Apple M3 Ultra (512 GB) Q6_K 98.3 GB / 384 GB 7.7 tok/s Runs well
57 ptsUsable (C) · 57 of 100 points: Model quality 28.0/45, Memory fit 10.0/25, Speed 4.9/15, Memory use 15.0/15 NVIDIA GeForce RTX 5090 Q3_K_M 43.3 GB / 32.0 GB 4.2 tok/s Runs with offload

Check it on your own hardware

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.