LLMs · Alibaba
Qwen2 72B Instruct
- Parameters
- 73B
- Architecture
- Dense
- Max context
- 32K tokens
- License
- qwen
- Released
- 2024-05
How to run it
Quickest · Ollama
One command downloads a ready-to-run version and starts a chat.
ollama run qwen2:72bQuantized files · GGUF
Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.
huggingface.co/Qwen/Qwen2-72B-Instruct-GGUF (opens in a new tab)
ollama run hf.co/Qwen/Qwen2-72B-Instruct-GGUFOriginal weights · Hugging Face
Full-precision files from Alibaba, for fine-tuning, vLLM, Transformers.
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 | 29.1 GB | 32.4 GB | 40.4 GB | GGUF (opens in a new tab) |
Q3_K_M |
Low | 35.6 GB | 38.9 GB | 47.0 GB | GGUF (opens in a new tab) |
Q4_K_M |
Good | 44.4 GB | 47.6 GB | 55.7 GB | GGUF (opens in a new tab) |
Q5_K_M |
High | 51.6 GB | 54.9 GB | 63.0 GB | GGUF · part 1 of 2 (opens in a new tab) |
Q6_K |
High | 59.6 GB | 62.9 GB | 71.0 GB | GGUF · part 1 of 2 (opens in a new tab) |
Q8_0 |
Very high | 77.1 GB | 80.4 GB | 88.4 GB | GGUF · part 1 of 2 (opens in a new tab) |
F16 |
Full | 145 GB | 149 GB | 157 GB | GGUF · part 1 of 4 (opens in a new tab) Original weights (opens in a new tab) |
Best hardware for Qwen2 72B Instruct
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 |
|---|---|---|---|---|---|
| 78 ptsGreat (A) · 78 of 100 points: Model quality 27.4/45, Memory fit 25.0/25, Speed 13.0/15, Memory use 12.6/15 | NVIDIA RTX PRO 6000 Blackwell | Q4_K_M | 52.4 GB / 96.0 GB | 30.3 tok/s | Runs well |
| 74 ptsGreat (A) · 74 of 100 points: Model quality 28.5/45, Memory fit 25.0/25, Speed 5.6/15, Memory use 15.0/15 | Apple M4 Max (128 GB) | Q6_K | 70.6 GB / 96.0 GB | 5.5 tok/s | Runs well |
| 72 ptsGreat (A) · 72 of 100 points: Model quality 28.5/45, Memory fit 25.0/25, Speed 3.8/15, Memory use 15.0/15 | AMD Ryzen AI Max+ 395 (128 GB) | Q6_K | 70.6 GB / 96.0 GB | 2.6 tok/s | Runs well |
| 72 ptsGreat (A) · 72 of 100 points: Model quality 28.5/45, Memory fit 25.0/25, Speed 4.2/15, Memory use 14.4/15 | NVIDIA GB10 (128 GB) | Q6_K | 72.1 GB / 115 GB | 3.2 tok/s | Runs well |
| 70 ptsGood (B) · 70 of 100 points: Model quality 24.2/45, Memory fit 24.0/25, Speed 7.6/15, Memory use 15.0/15 | Apple M4 Max (64 GB) | Q3_K_M | 42.8 GB / 48.0 GB | 10.3 tok/s | Tight fit |
| 69 ptsGood (B) · 69 of 100 points: Model quality 27.4/45, Memory fit 25.0/25, Speed 8.1/15, Memory use 9.5/15 | Apple M2 Ultra (192 GB) | Q4_K_M | 59.2 GB / 144 GB | 11.6 tok/s | Runs well |
| 69 ptsGood (B) · 69 of 100 points: Model quality 24.2/45, Memory fit 24.0/25, Speed 6.5/15, Memory use 15.0/15 | Apple M1 Max (64 GB) | Q3_K_M | 42.8 GB / 48.0 GB | 7.5 tok/s | Tight fit |
| 68 ptsGood (B) · 68 of 100 points: Model quality 27.4/45, Memory fit 25.0/25, Speed 8.2/15, Memory use 7.6/15 | Apple M3 Ultra (256 GB) | Q4_K_M | 63.0 GB / 192 GB | 11.9 tok/s | Runs well |
| 65 ptsGood (B) · 65 of 100 points: Model quality 27.4/45, Memory fit 25.0/25, Speed 8.2/15, Memory use 4.7/15 | Apple M3 Ultra (512 GB) | Q4_K_M | 78.4 GB / 384 GB | 11.9 tok/s | Runs well |
| 65 ptsGood (B) · 65 of 100 points: Model quality 20.2/45, Memory fit 24.0/25, Speed 6.1/15, Memory use 15.0/15 | Apple M4 Pro (48 GB) | Q2_K | 35.2 GB / 36.0 GB | 6.5 tok/s | Tight fit |
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.