Qwen
Qwen2 72B Instruct
Qwen2 72B Instruct is an open-weights 72B-parameter model by Qwen. It is listed as in-progress on the local.ai leaderboard.
open weights
Parameters
72B
Benchmarks (local.ai snapshot)
HuggingFace metadata · updated 2024-10-08
Metadata refreshed from the HuggingFace API (snapshot 2026-08-15T15:53:42.405Z). Context length and long-form descriptions are the next pipeline stage.
Does it fit your hardware?
Detect or select your GPU first. Green = weights + context headroom, yellow = tight, grey = does not fit.
Q4_K_M 41.8GQ6_K 64.8GQ8_0 79.2GFP16 144G| Quantization | ≈ File size | Note |
|---|---|---|
| IQ2_XXS | 19.4 GB | Extreme compression, largest quality loss |
| IQ3_XXS | 25.9 GB | Very small |
| Q3_K_M | 33.1 GB | Small |
| NVFP4 | 36 GB | NVIDIA 4-bit |
| Q4_K_S | 38.9 GB | Common |
| Q4_K_M | 41.8 GB | Most common sweet spot |
| Q5_K_M | 54.7 GB | Balanced |
| Q6_K | 64.8 GB | Near-lossless |
| Q8_0 | 79.2 GB | Near-lossless |
| FP16 | 144 GB | Original weights |
Actual files on HuggingFace
| File | Size |
|---|---|
| model-00006-of-00037.safetensors | 3.72 GB |
| model-00010-of-00037.safetensors | 3.72 GB |
| model-00014-of-00037.safetensors | 3.72 GB |
| model-00018-of-00037.safetensors | 3.72 GB |
| model-00022-of-00037.safetensors | 3.72 GB |
| model-00026-of-00037.safetensors | 3.72 GB |
GGUF: bartowski/Qwen2.5-72B-Instruct-GGUF · 65,924 downloads
| File | Size | |
|---|---|---|
| Qwen2.5-72B-Instruct-IQ1_M.gguf | 22.11 GB | download |
| Qwen2.5-72B-Instruct-IQ2_M.gguf | 27.32 GB | download |
| Qwen2.5-72B-Instruct-IQ2_XS.gguf | 25.20 GB | download |
| Qwen2.5-72B-Instruct-IQ2_XXS.gguf | 23.74 GB | download |
| Qwen2.5-72B-Instruct-IQ3_M.gguf | 33.07 GB | download |
| Qwen2.5-72B-Instruct-IQ3_XXS.gguf | 29.66 GB | download |
| Qwen2.5-72B-Instruct-IQ4_XS.gguf | 36.98 GB | download |
| Qwen2.5-72B-Instruct-Q2_K.gguf | 27.76 GB | download |
| Qwen2.5-72B-Instruct-Q2_K_L.gguf | 28.90 GB | download |
| Qwen2.5-72B-Instruct-Q3_K_L.gguf | 36.79 GB | download |
| Qwen2.5-72B-Instruct-Q3_K_M.gguf | 35.11 GB | download |
| Qwen2.5-72B-Instruct-Q3_K_S.gguf | 32.12 GB | download |
GGUF: mradermacher/Qwen2.5-72B-Instruct-heretic-i1-GGUF · 9,327 downloads
| File | Size | |
|---|---|---|
| Qwen2.5-72B-Instruct-heretic.i1-IQ1_M.gguf | 22.11 GB | download |
| Qwen2.5-72B-Instruct-heretic.i1-IQ1_S.gguf | 21.13 GB | download |
| Qwen2.5-72B-Instruct-heretic.i1-IQ2_M.gguf | 27.32 GB | download |
| Qwen2.5-72B-Instruct-heretic.i1-IQ2_S.gguf | 26.02 GB | download |
| Qwen2.5-72B-Instruct-heretic.i1-IQ2_XS.gguf | 25.20 GB | download |
| Qwen2.5-72B-Instruct-heretic.i1-IQ2_XXS.gguf | 23.74 GB | download |
| Qwen2.5-72B-Instruct-heretic.i1-IQ3_M.gguf | 33.07 GB | download |
| Qwen2.5-72B-Instruct-heretic.i1-IQ3_S.gguf | 32.12 GB | download |
| Qwen2.5-72B-Instruct-heretic.i1-IQ3_XS.gguf | 30.59 GB | download |
| Qwen2.5-72B-Instruct-heretic.i1-IQ3_XXS.gguf | 29.66 GB | download |
| Qwen2.5-72B-Instruct-heretic.i1-IQ4_XS.gguf | 36.98 GB | download |
| Qwen2.5-72B-Instruct-heretic.i1-Q2_K.gguf | 27.76 GB | download |
What you save self-hosting this model
Uses the model's own API price when it exists; otherwise the closest commercial equivalent.
Waiting for hardware detection…
RunLocal presents third-party leaderboard data (local.ai / Exo Labs, 2026-08) with its own fit and cost estimates. Verify pricing, licenses and file sizes at the source before deploying a model.