D J Lougen Q1_S 35B Parameters

VRAM Requirements for
hermes-qwen3.5-35b-a3b-GGUF Q1_S

To run hermes-qwen3.5-35b-a3b-GGUF locally at Q1_S quantization, you need at minimum 37.57 GB of GPU VRAM.

37.57 GB
Required VRAM
35 GB
File Size
8K tokens
Context Window
35B
Parameters
Estimated VRAM Required
37.57
GB
Prosumer / Workstation
0 16GB
RTX 3080 48GB
A6000 80GB+

Recommended GPU Configurations

Budget $1,200 – $1,800

2× RTX 3090 (48GB total)

48 GB

Dual-GPU via tensor parallelism. Best cost per GB at this tier.

Balanced $4,000 – $5,500

NVIDIA A6000 (48GB)

48 GB

Single-card 48GB pro GPU. Clean setup, no multi-GPU overhead.

Ultimate $8,000 – $12,000

NVIDIA A100 40GB SXM

40 GB HBM2e

Data-centre HBM2e bandwidth. Dramatically faster throughput.

📊 VRAM Calculation Breakdown

Model File Size (Q1_S) 35 GB
Context Overhead (8,192 tokens × 35B × 2 ÷ 1M) 0.573 GB
System Buffer (OS + CUDA runtime) 2.00 GB
Total Required VRAM 37.57 GB

Try a Different Quantization

Use the interactive calculator to compare hermes-qwen3.5-35b-a3b-GGUF across all available formats.

Open Live Calculator →

hermes-qwen3.5-35b-a3b-GGUF — Other Quantizations

Q2_M 10.94 GB file
13.51 GB
Q2_S 10.94 GB file
13.51 GB
Q2_XXS 10.94 GB file
13.51 GB
Q3_K_M 15.31 GB file
17.88 GB
Q3_K_S 15.31 GB file
17.88 GB
Q3_M 15.31 GB file
17.88 GB
Q3_S 15.31 GB file
17.88 GB
Q3_XXS 15.31 GB file
17.88 GB
Q4_K_M 19.69 GB file
22.26 GB
Q4_K_S 19.69 GB file
22.26 GB
Q4_NL 19.69 GB file
22.26 GB
Q4_XS 19.69 GB file
22.26 GB
Q5_K_M 24.06 GB file
26.63 GB
Q5_K_S 24.06 GB file
26.63 GB
Q6_K 28.44 GB file
31.01 GB
Q1_S Current
37.57 GB
Q1_M 35 GB file
37.57 GB
Q8_0 37.19 GB file
39.76 GB
Advertisement Zone

Frequently Asked Questions

Can I run hermes-qwen3.5-35b-a3b-GGUF Q1_S on a consumer GPU?
Running hermes-qwen3.5-35b-a3b-GGUF Q1_S locally requires 37.57 GB VRAM, which exceeds consumer GPUs. You'll need prosumer cards like the NVIDIA A6000 (48GB) or an A100 (80GB).
What happens if I don't have enough VRAM?
If your GPU VRAM is insufficient, llama.cpp and similar tools will offload model layers to system RAM (CPU inference). This is much slower — expect 10-50× the generation latency compared to full GPU inference.
Can I use multiple GPUs to run hermes-qwen3.5-35b-a3b-GGUF?
Yes! Tools like llama.cpp, vLLM, and Ollama support tensor parallelism across multiple GPUs. For example, 2× RTX 3090 (24GB each) gives you 48GB total VRAM, which can run many large models.
Is Q1_S quality good enough for production?
Q1_S is suitable for specialized use cases. Check community benchmarks for specific quality metrics.