Hermes-3-Llama-3.1-70B GGUF size and VRAM requirements
bartowski/Hermes-3-Llama-3.1-70B-GGUF is a very large language model with 70.55 billion parameters, built on the llama architecture. It is released under the llama3 license and has been downloaded 134,649 times.
Under the hood it uses 80 transformer layers, a hidden size of 8,192, 64 attention heads. It uses grouped-query attention (64 query heads sharing 8 key/value heads), which already trims KV-cache memory compared with full multi-head attention.
To run bartowski/Hermes-3-Llama-3.1-70B-GGUF locally at a 4,096-token context, its quantized versions need between 17.65 GB (IQ1_M, lowest quality) and 71.88 GB (Q8_0, highest quality) of memory, weights plus KV cache and a system margin included. Context length drives that memory directly: at 4,096 tokens the KV cache for Hermes-3-Llama-3.1-70B-GGUF is about 1.25 GB, rising to roughly 40.0 GB at its full 131,072-token context. Shorter prompts free up memory for a higher-quality quantization.
For most users the best balance is IQ2_XS, needing about 21.74 GB. That means bartowski/Hermes-3-Llama-3.1-70B-GGUF fits entirely in the VRAM of a 24 GB GPU or larger, running fully on the GPU.
Available GGUF quantizations for bartowski/Hermes-3-Llama-3.1-70B-GGUF include IQ1_M, IQ2_XXS, IQ2_XS, IQ2_M, Q2_K, Q2_K_L, IQ3_XXS, Q3_K_S, IQ3_M, Q3_K_M, Q3_K_L, IQ4_XS, Q3_K_XL, Q4_K_S, Q4_K_M, Q4_K_L, Q5_K_M, Q6_K, Q8_0. The model supports a native context length of up to 131,072 tokens; a longer context grows the KV cache and the memory needed.
GGUF file size and memory by quantization
Compare real GGUF weight sizes, estimated KV cache and total memory for Q4, Q5, Q8 and every quantization published in this repository.
| Quant. | Bits | Quality | Weights | KV | Total | Speed~ | Verdict |
|---|---|---|---|---|---|---|---|
| IQ1_M | 1.9 | Very low | 15.6 GB | 1.25 GB | 17.65 GB | 3.2 t/s | Offload |
| IQ2_XXS | 2.17 | Very low | 17.79 GB | 1.25 GB | 19.84 GB | 2.8 t/s | Offload |
| IQ2_XS | 2.4 | Very low | 19.69 GB | 1.25 GB | 21.74 GB | 2.5 t/s | Offload |
| IQ2_M | 2.73 | Low | 22.46 GB | 1.25 GB | 24.51 GB | — | Insufficient |
| Q2_K | 2.99 | Low | 24.56 GB | 1.25 GB | 26.61 GB | — | Insufficient |
| Q2_K_L | 3.11 | Low | 25.52 GB | 1.25 GB | 27.57 GB | — | Insufficient |
| IQ3_XXS | 3.11 | Low | 25.58 GB | 1.25 GB | 27.63 GB | — | Insufficient |
| Q3_K_S | 3.51 | Fair | 28.79 GB | 1.25 GB | 30.84 GB | — | Insufficient |
| IQ3_M | 3.62 | Fair | 29.74 GB | 1.25 GB | 31.79 GB | — | Insufficient |
| Q3_K_M | 3.89 | Fair | 31.91 GB | 1.25 GB | 33.96 GB | — | Insufficient |
| Q3_K_L | 4.21 | Good | 34.59 GB | 1.25 GB | 36.64 GB | — | Insufficient |
| IQ4_XS | 4.3 | Good | 35.3 GB | 1.25 GB | 37.35 GB | — | Insufficient |
| Q3_K_XL | 4.32 | Good | 35.45 GB | 1.25 GB | 37.5 GB | — | Insufficient |
| Q4_K_S | 4.57 | Good | 37.58 GB | 1.25 GB | 39.63 GB | — | Insufficient |
| Q4_K_M | 4.82 | Good | 39.6 GB | 1.25 GB | 41.65 GB | — | Insufficient |
| Q4_K_L | 4.91 | Good | 40.33 GB | 1.25 GB | 42.38 GB | — | Insufficient |
| Q5_K_M | 5.66 | Very good | 46.52 GB | 1.25 GB | 48.57 GB | — | Insufficient |
| Q6_K | 6.56 | Excellent | 53.91 GB | 1.25 GB | 55.96 GB | — | Insufficient |
| Q8_0 | 8.5 | Excellent | 69.83 GB | 1.25 GB | 71.88 GB | — | Insufficient |
KV cache computed from the model's exact architecture. Speed is a rough estimate bounded by memory bandwidth.
Frequently asked questions
What kind of model is bartowski/Hermes-3-Llama-3.1-70B-GGUF?
bartowski/Hermes-3-Llama-3.1-70B-GGUF is a language model with 70.55 billion parameters, based on the llama architecture. It is released under the llama3 license and distributed as GGUF files for local inference.
How much VRAM do you need to run bartowski/Hermes-3-Llama-3.1-70B-GGUF?
You need about 21.74 GB of VRAM to run bartowski/Hermes-3-Llama-3.1-70B-GGUF entirely on the GPU using the IQ2_XS quantization (at a 4,096-token context). Smaller quantizations lower the requirement at the cost of quality.
Can I run bartowski/Hermes-3-Llama-3.1-70B-GGUF on an 8 GB GPU?
Partially. bartowski/Hermes-3-Llama-3.1-70B-GGUF only fits on an 8 GB GPU by offloading part of it to system RAM (with IQ2_XS), which runs but is slower.
Can I run bartowski/Hermes-3-Llama-3.1-70B-GGUF on a 16 GB GPU?
Partially. bartowski/Hermes-3-Llama-3.1-70B-GGUF only fits on a 16 GB GPU by offloading part of it to system RAM (with Q4_K_L), which runs but is slower.
Can I run bartowski/Hermes-3-Llama-3.1-70B-GGUF on a 24 GB GPU?
Yes. With 24 GB of VRAM you can run bartowski/Hermes-3-Llama-3.1-70B-GGUF fully on the GPU using IQ2_XS (about 21.74 GB).
What context length does bartowski/Hermes-3-Llama-3.1-70B-GGUF support?
bartowski/Hermes-3-Llama-3.1-70B-GGUF supports a native context length of up to 131,072 tokens. A longer context grows the KV cache, so it increases the memory needed to run the model.
What is the best quantization for bartowski/Hermes-3-Llama-3.1-70B-GGUF?
For bartowski/Hermes-3-Llama-3.1-70B-GGUF, a strong default is Q4_K_M, which needs about 41.65 GB and keeps most of the quality while roughly halving the memory versus 8-bit. With VRAM to spare, Q5_K_M or Q6_K add a little more quality; if you are tight on memory, a smaller quantization still runs. Pick the highest quantization that fits your VRAM.