Arete 27B — Hardware Requirements & GPU Compatibility
ChatFunctions
Arete-27B is a 27 billion parameter model from Nous Research, built by fine-tuning Alibaba's Qwen3.8-27B to shift its answers on politically sensitive topics away from the framing it learned from its Chinese training. Nous used supervised fine-tuning only, on about 20,000 prompts across seven value categories such as freedom of expression, democracy and data privacy, with answers distilled from the open Inkling model as teacher. There are no architectural changes, and the vision encoder was frozen. It is meant for general assistance and agentic work, and for research on how post-training changes model values. Fine-tuning used 64k context, and the model's maximum context length is 262,144 tokens. It is released under the Apache 2.0 license, which permits commercial use. At 27B parameters it runs on a single 24 GB consumer GPU or a well-equipped laptop once quantized.
Runs well on a 24 GB GPU
Runs from 12 GB (IQ3_XXS · 16K)Best on 24 GB (Q4_K_M · ~38 tok/s · 128K)Macs from 16 GB (IQ3_XS · ~12 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Arete 27B
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
ollama run hf.co/bartowski/NousResearch_Arete-27B-GGUF:Q4_K_MQ4_K_M fits fully on the RTX 4090 · 6.4 GB free · 16K context
Download: NousResearch_Arete-27B-Q4_K_M.gguf17.20 GBbartowski
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX 3090 (24 GB) (6 GB headroom) · Q4_K_M
- Generation speed
- ~36 tok/s
- generation speed
- Cost per 1M output tokens
- $1.43
- per 1M output tokens
How Much VRAM Does Arete 27B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 12.4 GB | 13.7 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 128K context · ~53 tok/s | |
| 12.7 GB | 14.1 GB | 3-bit small — 3-bit small quantization | Fits: 128K context · ~52 tok/s | |
| 14.1 GB | 15.4 GB | 3-bit medium — 3-bit medium quantization | Fits: 128K context · ~47 tok/s | |
| 14.4 GB | 15.8 GB | 4-bit · legacy — 4-bit legacy quantization | Fits: 128K context · ~45 tok/s | |
| 17.2 GB | 18.5 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 128K context · ~38 tok/s | |
| 20.2 GB | 21.6 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 64K context · ~32 tok/s | |
| 23.3 GB | 24.7 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 16K context · ~28 tok/s | |
| 28.1 GB | 29.4 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Does not fit: needs 28.6 GB |
Selected Q4_K_M
4-bit medium quantization17.6GBat 16K context
- Weights
- 16.4 GB
- KV cache
- 0.9 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 6.4 GB free · ~38 tok/s
≥50 tok/s needs ≥1.32 TB/s · RTX 5090 ~68
Which Hardware Can Run Arete 27B?
Q4_K_M · 17.2 GBArete 27B (Q4_K_M) requires 17.2 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 23+ GB is recommended. Using the full 262K context window can add up to 13.9 GB, bringing total usage to 31.0 GB. 8 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 122 devices with unified memory can run Arete 27B, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Mini M5 Pro (24 GB).
20 GB
fits · 2.4 GB free · up to 32K ctx
1 GPU
RX 7900 XT~28Simulate output on RX 7900 XT32–40 GB
fits · 14.4 GB free · up to 256K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~68Simulate output on RTX 5090M5 Max36 GB~19Simulate output on Mac Studio M5 Max (36 GB)M532 GB~6Simulate output on MacBook Air 13" M5 (32 GB)Orin~8Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~5Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Qwen3.8 27BAlibaba | 122.0 | 17.9 GB | ~38 | same size |
| Qwen3.6 27BAlibaba | 111.2 | 17.9 GB | ~38 | same size |
| Qwen3.5 27BAlibaba | 103.8 | 17.9 GB | ~38 | same size |
| Gemma 3 27B ITGoogle | 79.4 | 21.7 GB | ~36 | same size |
| Gemma 4 26B A4BGoogle | — | 16.5 GB | ~171 | same size |
| ERNIE 4.5 VL 28B A3B PTBaidu | — | 60 GB | —needs > 24 GB | same size |
Specifications
- Publisher
- Nous Research
- Parameters
- 27.4B
- Architecture
- Qwen3_5ForConditionalGeneration
- Context Length
- 262,144 tokens
- Vocabulary Size
- 248,320
- Release Date
- 2026-11-06
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does Arete 27B need?
Arete 27B requires 17.2 GB of VRAM at Q4_K_M, or 55.5 GB at BF16. Full 262K context adds up to 13.9 GB (31.0 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 27.4B × 4.8 bits ÷ 8 = 16.4 GB
KV Cache + Overhead ≈ 0.8 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 ≈ 14.6 GB (at full 262K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 4090 run Arete 27B?
Yes, at Q6_K (23.3 GB) or lower. Higher quantizations like Q8_0 (28.1 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
What's the best quantization for Arete 27B?
For Arete 27B, Q4_K_M (17.2 GB) offers the best balance of quality and VRAM usage. Q4_K_L (17.5 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 8.3 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Arete 27B on a Mac?
Yes — Mac Mini M2 Pro (16 GB) and 133 other Macs can run Arete 27B, the smallest at IQ3_XS. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 17.2 GB of usable unified memory (RAM minus macOS overhead), from Mac Mini M5 Pro (24 GB).
Can I run Arete 27B locally?
Yes — Arete 27B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Arete 27B?
At Q4_K_M, Arete 27B can reach ~280 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~38 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 17.16 × 0.65 = ~303 tok/s
Estimated speed at Q4_K_M (17.2 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Arete 27B?
At Q4_K_M, the download is about 16.41 GB. The full-precision BF16 version is 54.71 GB. The smallest option (IQ2_XXS) is 7.52 GB.
Which GPUs can run Arete 27B?
8 consumer GPUs can run Arete 27B at Q4_K_M. Top options include NVIDIA GeForce RTX 5090, AMD Radeon RX 7900 XT, AMD Radeon RX 7900 XTX. 1 GPU has plenty of headroom for comfortable inference.
Which devices can run Arete 27B?
122 devices with unified memory can run Arete 27B at Q4_K_M (17.2 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.