Meta
Llama 3.3 70B Instruct
Llama 3.3 70B Instruct is an open-weights 70B-parameter model by Meta. It is listed as in-progress on the local.ai leaderboard.
open weights
Parameters
70B
Benchmarks (local.ai snapshot)
HuggingFace metadata · updated 2024-12-21
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 40.6GQ6_K 63GQ8_0 77GFP16 140G| Quantization | ≈ File size | Note |
|---|---|---|
| IQ2_XXS | 18.9 GB | Extreme compression, largest quality loss |
| IQ3_XXS | 25.2 GB | Very small |
| Q3_K_M | 32.2 GB | Small |
| NVFP4 | 35 GB | NVIDIA 4-bit |
| Q4_K_S | 37.8 GB | Common |
| Q4_K_M | 40.6 GB | Most common sweet spot |
| Q5_K_M | 53.2 GB | Balanced |
| Q6_K | 63 GB | Near-lossless |
| Q8_0 | 77 GB | Near-lossless |
| FP16 | 140 GB | Original weights |
Actual files on HuggingFace
| File | Size |
|---|---|
| model-00008-of-00030.safetensors | 4.66 GB |
| model-00013-of-00030.safetensors | 4.66 GB |
| model-00018-of-00030.safetensors | 4.66 GB |
| model-00023-of-00030.safetensors | 4.66 GB |
| model-00028-of-00030.safetensors | 4.66 GB |
| model-00003-of-00030.safetensors | 4.66 GB |
GGUF: MaziyarPanahi/Llama-3.3-70B-Instruct-GGUF · 169,345 downloads
| File | Size | |
|---|---|---|
| Llama-3.3-70B-Instruct.Q2_K.gguf | 24.56 GB | download |
| Llama-3.3-70B-Instruct.Q3_K_L.gguf | 34.59 GB | download |
| Llama-3.3-70B-Instruct.Q3_K_M.gguf | 31.91 GB | download |
| Llama-3.3-70B-Instruct.Q3_K_S.gguf | 28.79 GB | download |
| Llama-3.3-70B-Instruct.Q4_K_M.gguf | 39.60 GB | download |
| Llama-3.3-70B-Instruct.Q4_K_S.gguf | 37.58 GB | download |
| Llama-3.3-70B-Instruct.Q5_K_M.gguf | 46.52 GB | download |
| Llama-3.3-70B-Instruct.Q5_K_S.gguf | 45.32 GB | download |
| Llama-3.3-70B-Instruct.Q6_K.gguf-00001-of-00006.gguf | 10.59 GB | download |
| Llama-3.3-70B-Instruct.Q6_K.gguf-00002-of-00006.gguf | 9.33 GB | download |
| Llama-3.3-70B-Instruct.Q6_K.gguf-00003-of-00006.gguf | 9.16 GB | download |
| Llama-3.3-70B-Instruct.Q6_K.gguf-00004-of-00006.gguf | 9.26 GB | download |
GGUF: bartowski/Llama-3.3-70B-Instruct-abliterated-GGUF · 61,272 downloads
| File | Size | |
|---|---|---|
| Llama-3.3-70B-Instruct-abliterated-IQ1_M.gguf | 15.60 GB | download |
| Llama-3.3-70B-Instruct-abliterated-IQ2_M.gguf | 22.46 GB | download |
| Llama-3.3-70B-Instruct-abliterated-IQ2_S.gguf | 20.71 GB | download |
| Llama-3.3-70B-Instruct-abliterated-IQ2_XS.gguf | 19.69 GB | download |
| Llama-3.3-70B-Instruct-abliterated-IQ2_XXS.gguf | 17.79 GB | download |
| Llama-3.3-70B-Instruct-abliterated-IQ3_M.gguf | 29.74 GB | download |
| Llama-3.3-70B-Instruct-abliterated-IQ3_XXS.gguf | 25.58 GB | download |
| Llama-3.3-70B-Instruct-abliterated-IQ4_NL.gguf | 37.30 GB | download |
| Llama-3.3-70B-Instruct-abliterated-IQ4_XS.gguf | 35.30 GB | download |
| Llama-3.3-70B-Instruct-abliterated-Q2_K.gguf | 24.56 GB | download |
| Llama-3.3-70B-Instruct-abliterated-Q2_K_L.gguf | 25.52 GB | download |
| Llama-3.3-70B-Instruct-abliterated-Q3_K_L.gguf | 34.59 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.