Uyu 2 28B — Hardware Requirements & GPU Compatibility
ChatRoleplay
Uyu 2 28B is a 28.2B-parameter open language model from mente-ai. It supports a context window of up to 262,144 tokens. At Q4_K_M it needs about 18.53 GB of VRAM — see which GPUs and Macs can run it below.
Runs well on a 24 GB GPU
Runs on 24 GB (Q4_K_M · ~35 tok/s · 8K)Macs from 24 GB (Q4_K_M · ~12 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Uyu 2 28B
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
ollama run hf.co/mente-ai/uyu-2-28B-GGUFQ4_K_M needs offload on the RTX 4090 · 3.8 GB short · 16K context
Download: uyu-2-28b-Q4_K_M.gguf17.21 GBmente-ai
Other ways
each opens step-by-stepRun in cloud
Fits on RTX A6000 (48 GB) (20 GB headroom) · Q4_K_M
- Generation speed
- ~27 tok/s
- generation speed
- Cost per 1M output tokens
- $3.41
- per 1M output tokens
How Much VRAM Does Uyu 2 28B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 13.6 GB | 33.4 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 16K context · ~48 tok/s | |
| 15.4 GB | 35.2 GB | 3-bit medium — 3-bit medium quantization | Fits: 8K context · ~43 tok/s | |
| 18.5 GB | 38.4 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 8K context · ~35 tok/s | |
| 21.7 GB | 41.5 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 4K context · ~30 tok/s | |
| 24.9 GB | 44.7 GB | 6-bit, very good — 6-bit quantization, very good quality | Does not fit: needs 34.1 GB | |
| 29.8 GB | 49.6 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Does not fit: needs 39.1 GB | |
| 58.0 GB | 77.8 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Does not fit: needs 67.2 GB |
Selected Q4_K_M
4-bit medium quantization27.8GBat 16K context
- Weights
- 16.9 GB
- KV cache
- 10.6 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · fits up to 8K context · ~35 tok/s
≥50 tok/s needs ≥1.42 TB/s · RTX 5090 ~63
Which Hardware Can Run Uyu 2 28B?
Q4_K_M · 18.5 GBUyu 2 28B (Q4_K_M) requires 18.5 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 25+ GB is recommended. Using the full 262K context window can add up to 167.8 GB, bringing total usage to 186.3 GB. 8 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 122 devices with unified memory can run Uyu 2 28B, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Mini M5 Pro (24 GB).
20 GB
tight · up to 4K ctx
1 GPU
RX 7900 XT~26Simulate output on RX 7900 XT32–40 GB
fits · 4.2 GB free · up to 16K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~63Simulate output on RTX 5090M5 Max36 GB~17Simulate output on Mac Studio M5 Max (36 GB)M532 GB~6Simulate output on MacBook Air 13" M5 (32 GB)Orin~7Simulate 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 | 122.1 | 17.9 GB | ~38 | same size |
| Qwen3.6 27BAlibaba | 111.2 | 17.9 GB | ~38 | same size |
| Hy Memory 30B A3BTencent | — | 19.1 GB | ~184 | same size |
| Muse Glimmer 30BMeta | — | 19.6 GB | ~36 | same size |
| Hy MT2 30B A3BTencent | — | 19.1 GB | ~170 | same size |
| Gemma 4 26B A4BGoogle | — | 16.9 GB | ~153 | same size |
Specifications
- Publisher
- mente-ai
- Parameters
- 28.2B
- Architecture
- Uyu2ForCausalLM
- Context Length
- 262,144 tokens
- Vocabulary Size
- 262,144
- Release Date
- 2026-07-10
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does Uyu 2 28B need?
Uyu 2 28B requires 18.5 GB of VRAM at Q4_K_M, or 58.0 GB at BF16. Full 262K context adds up to 167.8 GB (186.3 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 28.2B × 4.8 bits ÷ 8 = 16.9 GB
KV Cache + Overhead ≈ 1.6 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 ≈ 169.4 GB (at full 262K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 4090 run Uyu 2 28B?
Yes, at Q5_K_M (21.7 GB) or lower. Higher quantizations like Q6_K (24.9 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
What's the best quantization for Uyu 2 28B?
For Uyu 2 28B, Q4_K_M (18.5 GB) offers the best balance of quality and VRAM usage. Q5_K_M (21.7 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 13.6 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Uyu 2 28B on a Mac?
Yes — Mac Mini M5 Pro (24 GB) and 106 other Macs can run Uyu 2 28B. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 18.5 GB of usable unified memory (RAM minus macOS overhead).
Can I run Uyu 2 28B locally?
Yes — Uyu 2 28B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Uyu 2 28B?
At Q4_K_M, Uyu 2 28B can reach ~259 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~35 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 ÷ 18.5 × 0.65 = ~281 tok/s
Estimated speed at Q4_K_M (18.5 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Uyu 2 28B?
At Q4_K_M, the download is about 16.91 GB. The full-precision BF16 version is 56.36 GB. The smallest option (Q2_K) is 11.98 GB.
Which GPUs can run Uyu 2 28B?
8 consumer GPUs can run Uyu 2 28B 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 Uyu 2 28B?
122 devices with unified memory can run Uyu 2 28B at Q4_K_M (18.5 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.