Hermes 4 405B — Hardware Requirements & GPU Compatibility
ChatReasoningRoleplay
Hermes 4 405B is a 405.9B-parameter open language model from Nous Research in the Hermes family. It supports a context window of up to 131,072 tokens. At Q4_K_M it needs about 244.87 GB of VRAM — see which GPUs and Macs can run it below.
Hermes 4 405B runs on a 256 GB+ Mac or a datacenter GPU
Best local pick → Mac Studio (M3 Ultra, 256GB) · Q4_K_M · 252.3 GB · ~2 tok/s
Macs from 192 GB (Q2_K · ~3 tok/s)
Llama 3 Community
Not set up for it?
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Hermes 4 405B
One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.
Recommended
vllm serve NousResearch/Hermes-4-405B --max-model-len 16384 --tensor-parallel-size 8vLLM repo with its own quantisation
Download: Weights on Hugging Face
Ways to runOther ways
each opens step-by-stepRun in cloud
Rent 4× H100 SXM · Q4_K_M
- Estimated hourly price
- ~$10.76/hr
- est., whole server
- GPUs
- 4 GPUs
- 320 GB of VRAM
$5+ bonus for new users
How Much VRAM Does Hermes 4 405B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | Runs on | |
|---|---|---|---|---|
| 173.8 GB | 189.7 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Runs on B200 192 GB | |
| 199.2 GB | 215.1 GB | 3-bit medium — 3-bit medium quantization | Runs on Instinct MI325X 256 GB | |
| 244.9 GB | 260.7 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Runs on Instinct MI325X 256 GB | |
| 290.5 GB | 306.4 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Runs on 5× H100 SXM 80 GB | |
| 336.2 GB | 352 GB | 6-bit, very good — 6-bit quantization, very good quality | Runs on 5× H100 SXM 80 GB | |
| 407.2 GB | 423.1 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Runs on 6× H100 SXM 80 GB | |
| 813.1 GB | 828.9 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Runs on 7× H200 SXM 141 GB |
Selected Q4_K_M
4-bit medium quantization252.3GBat 16K context
- Weights
- 243.5 GB
- KV cache
- 8.5 GB
- Overhead
- 0.3 GB
Server-class: ~252.3 GB at 16K context · one Instinct MI325X 256 GB
Which Hardware Can Run Hermes 4 405B?
Q4_K_M · 244.9 GBHermes 4 405B (Q4_K_M) requires 244.9 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 319+ GB is recommended. Using the full 131K context window can add up to 66.6 GB, bringing total usage to 311.5 GB. No consumer GPU has enough memory. 2 devices with unified memory can run Hermes 4 405B, including Mac Studio (M3 Ultra, 512GB), Mac Studio (M3 Ultra, 256GB).
192+ GB
Q4_K_M · best fit Q2_K (181.2 GB)
6 GPUs · 1 Mac class · 2 device families
B200~30Simulate output on B200B300~30Simulate output on B300Instinct MI350X~28Simulate output on Instinct MI350XM3 Ultra256–512 GB~3Simulate output on Mac Studio (M3 Ultra, 256GB)8x H100 SXM5~100Simulate output on NVIDIA DGX H1008x A100 SXM4~61Simulate output on NVIDIA DGX A100 640GB
By VRAM tier · typical GPUs and Macs at Q4_K_M
Consumer GPUs (up to 32 GB) and Macs up to 128 GB: too small for this model (needs ≥ 181.2 GB)8 GB tier →12 GB tier →16 GB tier →24 GB tier →Macs →
Alternatives to Hermes 4 405B
Hermes family · 4 models →Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Llama 4 Maverick 17B 128E InstructMeta | 87.1 | 272.7 GB | —needs > 24 GB | same size |
| Llama 3.1 405B InstructMeta | 83.5 | 275.6 GB | —needs > 24 GB | same size |
| MiniMax M3.1 FlashMiniMax | — | 258.6 GB | —needs > 24 GB | same size |
| Intern S2 397BInternLM | — | 242.6 GB | —needs > 24 GB | same size |
| Ornith 1.5 397Bornith-ai | — | 242.6 GB | —needs > 24 GB | same size |
| Trinity Large ThinkingArcee AI | — | 799.6 GB | —needs > 24 GB | same size |
Specifications
- Publisher
- Nous Research
- Family
- Hermes
- Parameters
- 405.9B
- Architecture
- LlamaForCausalLM
- Context Length
- 131,072 tokens
- Vocabulary Size
- 128,256
- Release Date
- 2025-08-06
- License
- Llama 3 Community
Frequently Asked Questions
How much VRAM does Hermes 4 405B need?
Hermes 4 405B requires 244.9 GB of VRAM at Q4_K_M, or 813.1 GB at BF16. Full 131K context adds up to 66.6 GB (311.5 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 405.9B × 4.8 bits ÷ 8 = 243.5 GB
KV Cache + Overhead ≈ 1.4 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
KV Cache + Overhead ≈ 68 GB (at full 131K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 5090 run Hermes 4 405B?
No — Hermes 4 405B requires at least 173.8 GB at Q2_K, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
What's the best quantization for Hermes 4 405B?
For Hermes 4 405B, Q4_K_M (244.9 GB) offers the best balance of quality and VRAM usage. Q5_K_M (290.5 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 173.8 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Hermes 4 405B on a Mac?
Yes — Mac Pro M2 Ultra (192 GB) and 3 other Macs can run Hermes 4 405B, the smallest at Q2_K. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 244.9 GB of usable unified memory (RAM minus macOS overhead), from Mac Studio (M3 Ultra, 256GB).
Can I run Hermes 4 405B locally?
Yes — Hermes 4 405B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Hermes 4 405B?
At Q4_K_M, Hermes 4 405B can reach ~20 tok/s on AMD Instinct MI350X. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B300 → 8000 ÷ 244.9 × 0.65 = ~21 tok/s
Estimated speed at Q4_K_M (244.9 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Hermes 4 405B?
At Q4_K_M, the download is about 243.51 GB. The full-precision BF16 version is 811.71 GB. The smallest option (Q2_K) is 172.49 GB.
Which GPUs can run Hermes 4 405B?
No single consumer GPU has enough VRAM to run Hermes 4 405B at Q4_K_M (244.9 GB). Multi-GPU or professional hardware is required.
Which devices can run Hermes 4 405B?
2 devices with unified memory can run Hermes 4 405B at Q4_K_M (244.9 GB), including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.