Qwen3.8 27B SLQ EAR98 — Hardware Requirements & GPU Compatibility
ChatFunctions
Qwen3.8 27B SLQ EAR98 is a 7.5B-parameter open language model from kaitchup in the Qwen 3.8 family. It supports a context window of up to 262,144 tokens. At Q4_K_M it needs about 5.24 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 32K)Best on 24 GB (Q4_K_M · ~125 tok/s · 256K)Macs from 8 GB (Q2_K · ~18 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Qwen3.8 27B SLQ EAR98
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 kaitchup/Qwen3.8-27B-SLQ-EAR98 --max-model-len 16384vLLM repo with its own quantisation
Download: Weights on Hugging Face
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX 3080 Ti (12 GB) (6 GB headroom) · Q4_K_M
- Generation speed
- ~113 tok/s
- generation speed
- Cost per 1M output tokens
- $0.30
- per 1M output tokens
How Much VRAM Does Qwen3.8 27B SLQ EAR98 Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 3.9 GB | 5.3 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 256K context · ~167 tok/s | |
| 4.4 GB | 5.7 GB | 3-bit medium — 3-bit medium quantization | Fits: 256K context · ~149 tok/s | |
| 5.2 GB | 6.6 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 256K context · ~125 tok/s | |
| 6.1 GB | 7.4 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 256K context · ~108 tok/s | |
| 6.9 GB | 8.3 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 256K context · ~95 tok/s | |
| 8.2 GB | 9.6 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 256K context · ~80 tok/s | |
| 15.7 GB | 17.1 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 128K context · ~42 tok/s |
Selected Q4_K_M
4-bit medium quantization5.7GBat 16K context
- Weights
- 4.5 GB
- KV cache
- 0.9 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 18.3 GB free · ~125 tok/s
≥50 tok/s needs ≥403 GB/s · RTX 5090 ~222, RTX 3090 Ti ~125
Which Hardware Can Run Qwen3.8 27B SLQ EAR98?
Q4_K_M · 5.2 GBQwen3.8 27B SLQ EAR98 (Q4_K_M) requires 5.2 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 7+ GB is recommended. Using the full 262K context window can add up to 13.9 GB, bringing total usage to 19.1 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 151 devices with unified memory can run Qwen3.8 27B SLQ EAR98, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 2.3 GB free · up to 32K ctx
16 GPUs · 1 device family
RTX 3080~94Simulate output on RTX 3080RTX 2080 Ti~76Simulate output on RTX 2080 TiRTX 3070 Ti~76Simulate output on RTX 3070 TiOrin Nano~13Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)20 GB
fits · 14.3 GB free · up to 256K ctx
1 GPU
RX 7900 XT~92Simulate output on RX 7900 XT32–40 GB
fits · 26.3 GB free · up to 256K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~222Simulate output on RTX 5090M5 Max36 GB~62Simulate output on Mac Studio M5 Max (36 GB)M532 GB~21Simulate output on MacBook Air 13" M5 (32 GB)Orin~25Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~17Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Alternatives to Qwen3.8 27B SLQ EAR98
Qwen 3.8 family · 12 models →Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Ling 3.0 TinyInclusion AI | — | 5.5 GB | ~364 | same size |
| BitCPM CANN 8BOpenBMB | — | 5.6 GB | ~127 | same size |
| Gemma 4 E4BGoogle | — | 5.6 GB | ~123 | same size |
| Olmo 3 7B InstructAllen AI | — | 8.4 GB | ~114 | same size |
| Phi Mini MoE InstructMicrosoft | — | 5.4 GB | ~239 | same size |
| Gemma 3n E4B ITGoogle | — | 7.7 GB | ~127 | same size |
Specifications
- Publisher
- kaitchup
- Family
- Qwen 3.8
- Parameters
- 7.5B
- Architecture
- Qwen3_5ForConditionalGeneration
- Context Length
- 262,144 tokens
- Vocabulary Size
- 248,320
- Release Date
- 2026-10-21
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does Qwen3.8 27B SLQ EAR98 need?
Qwen3.8 27B SLQ EAR98 requires 5.2 GB of VRAM at Q4_K_M, or 15.7 GB at BF16. Full 262K context adds up to 13.9 GB (19.1 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 7.5B × 4.8 bits ÷ 8 = 4.5 GB
KV Cache + Overhead ≈ 0.7 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
What's the best quantization for Qwen3.8 27B SLQ EAR98?
For Qwen3.8 27B SLQ EAR98, Q4_K_M (5.2 GB) offers the best balance of quality and VRAM usage. Q5_K_M (6.1 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 3.9 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Qwen3.8 27B SLQ EAR98 on a Mac?
Yes — Mac Mini M2 (8 GB) and 145 other Macs can run Qwen3.8 27B SLQ EAR98, 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 5.2 GB of usable unified memory (RAM minus macOS overhead), from Mac Mini M2 Pro (16 GB).
Can I run Qwen3.8 27B SLQ EAR98 locally?
Yes — Qwen3.8 27B SLQ EAR98 can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Qwen3.8 27B SLQ EAR98?
At Q4_K_M, Qwen3.8 27B SLQ EAR98 can reach ~916 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~125 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 ÷ 5.24 × 0.65 = ~992 tok/s
Estimated speed at Q4_K_M (5.2 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Qwen3.8 27B SLQ EAR98?
At Q4_K_M, the download is about 4.50 GB. The full-precision BF16 version is 14.99 GB. The smallest option (Q2_K) is 3.19 GB.
Which GPUs can run Qwen3.8 27B SLQ EAR98?
52 consumer GPUs can run Qwen3.8 27B SLQ EAR98 at Q4_K_M. Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 52 GPUs have plenty of headroom for comfortable inference.
Which devices can run Qwen3.8 27B SLQ EAR98?
151 devices with unified memory can run Qwen3.8 27B SLQ EAR98 at Q4_K_M (5.2 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.