Qwen 1 8B — Hardware Requirements & GPU Compatibility
Chat
Qwen-1.8B is Alibaba's first-generation 1.8-billion-parameter base language model, pretrained from scratch on over 2.2 trillion tokens of Chinese, English, multilingual, code, and math data, with the same roughly 150,000-token vocabulary used across the Qwen family. It is a raw pretrained model rather than a chat assistant; Alibaba's aligned Qwen-1.8B-Chat is built on top of it. Its main selling point is low-cost deployment: the card reports int4/int8 quantized versions needing under 2GB of memory for inference and as little as 6GB for fine-tuning, so it runs comfortably on almost any consumer GPU or even a laptop. Context length is 8,192 tokens. It is released under a custom Tongyi Qianwen Research License, free for academic research, with commercial use requiring direct contact with Alibaba. It was published in November 2023.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 8K)Best on 24 GB (Q4_K_M · ~542 tok/s · 8K)Macs from 8 GB (Q4_K_M · ~59 tok/s)
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Qwen 1 8B
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/mradermacher/Qwen-1_8B-GGUF:Q4_K_MQ4_K_M fits fully on the RTX 4090 · 22.1 GB free · 8K context
Download: Qwen-1_8B.Q4_K_M.gguf1.24 GBmradermacher
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX 3080 Ti (12 GB) (10 GB headroom) · Q4_K_M
- Generation speed
- ~490 tok/s
- generation speed
- Cost per 1M output tokens
- $0.069
- per 1M output tokens
How Much VRAM Does Qwen 1 8B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 0.9 GB | — | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 8K context · ~762 tok/s | |
| 0.9 GB | — | 3-bit small — 3-bit small quantization | Fits: 8K context · ~745 tok/s | |
| 1.0 GB | — | 3-bit medium — 3-bit medium quantization | Fits: 8K context · ~669 tok/s | |
| 1.0 GB | — | 4-bit · legacy — 4-bit legacy quantization | Fits: 8K context · ~649 tok/s | |
| 1.2 GB | — | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 8K context · ~542 tok/s | |
| 1.4 GB | — | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 8K context · ~455 tok/s | |
| 1.7 GB | — | 6-bit, very good — 6-bit quantization, very good quality | Fits: 8K context · ~392 tok/s | |
| 2.0 GB | — | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 8K context · ~324 tok/s |
Selected Q4_K_M
4-bit medium quantization1.9GBat 8K context
- Weights
- 0.7 GB
- KV cache
- 1.0 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 22.1 GB free · ~542 tok/s
≥50 tok/s needs ≥93 GB/s · RTX 5090 ~963, RTX 3090 Ti ~542
Which Hardware Can Run Qwen 1 8B?
Q4_K_M · 1.2 GBQwen 1 8B (Q4_K_M) requires 1.2 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 2+ GB is recommended. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 163 devices with unified memory can run Qwen 1 8B, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 6.1 GB free · up to 8K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~408Simulate output on RTX 3080RTX 2080 Ti~331Simulate output on RTX 2080 TiRTX 3070 Ti~327Simulate output on RTX 3070 TiM3*8 GB~59Simulate output on iMac 24" M3 (8 GB)Orin Nano~55Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~40Simulate output on iPhone 1720 GB
fits · 18.1 GB free · up to 8K ctx
1 GPU
RX 7900 XT~397Simulate output on RX 7900 XT32–40 GB
fits · 30.1 GB free · up to 8K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~963Simulate output on RTX 5090M5 Max36 GB~266Simulate output on Mac Studio M5 Max (36 GB)M532 GB~89Simulate output on MacBook Air 13" M5 (32 GB)Orin~110Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~73Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
How good is Qwen 1 8B?
llmrun Score blends public benchmark boards into one number; 100 = the average open model.
Not enough data yet for an llmrun Score.
A Score needs results on enough independent boards. How the Composite Score is built
Benchmarks
Here are all 6 boards that have results for this model, sorted by their weight in the llmrun Score; badges show who ran the test.
- MMLU
Epoch AI · Nov 2023Lab-reported
Lab-reported28.2%
#80/85 open
- BIG-Bench Hard
Epoch AI · Nov 2023Lab-reported
Lab-reported28.2%
#38/38 open
- GSM8K
Epoch AI · Nov 2023Lab-reported
Lab-reported21.2%
#54/62 open
- ARC Challenge
Epoch AI · Nov 2023Lab-reported
Lab-reported53.2%
#29/56 open
- BoolQ
Epoch AI · Nov 2023Lab-reported
Lab-reported68.0%
#28/38 open
- PIQA
Epoch AI · Nov 2023Lab-reported
Lab-reported73.3%
#36/40 open
Alternatives to Qwen 1 8B
Qwen family · 28 models →Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| D1 2BLiquid AI | — | 2.3 GB | ~423 | same size |
| LFM3 1.7BLiquid AI | — | 2.1 GB | ~455 | same size |
| Moondream2vikhyatk | — | 3 GB | ~516 | same size |
| JiRackUltra 1BCMSManhattan | — | 1.8 GB | ~461 | same size |
| Llm Jp 3.1 1.8B Instruct4llm-jp | — | 2.2 GB | ~360 | same size |
| DeepSeek R1 Distill Qwen 1.5BDeepSeek | — | 1.8 GB | ~461 | same size |
Specifications
- Publisher
- Alibaba
- Family
- Qwen
- Parameters
- 1.8B
- Architecture
- QWenLMHeadModel
- Context Length
- 8,192 tokens
- Vocabulary Size
- 151,936
- Release Date
- 2023-11-30
Frequently Asked Questions
How much VRAM does Qwen 1 8B need?
Qwen 1 8B requires 1.2 GB of VRAM at Q4_K_M, or 4.0 GB at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 1.8B × 4.8 bits ÷ 8 = 1.1 GB
KV Cache + Overhead ≈ 0.1 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.
VRAM usage by quantization
What's the best quantization for Qwen 1 8B?
For Qwen 1 8B, Q4_K_M (1.2 GB) offers the best balance of quality and VRAM usage. Q5_K_S (1.4 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 0.6 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Qwen 1 8B on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run Qwen 1 8B. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 1.2 GB of usable unified memory (RAM minus macOS overhead).
Can I run Qwen 1 8B locally?
Yes — Qwen 1 8B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Qwen 1 8B?
At Q4_K_M, Qwen 1 8B can reach ~3967 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~542 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 ÷ 1.21 × 0.65 = ~4298 tok/s
Estimated speed at Q4_K_M (1.2 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Qwen 1 8B?
At Q4_K_M, the download is about 1.10 GB. The full-precision BF16 version is 3.67 GB. The smallest option (IQ2_XXS) is 0.51 GB.
Which GPUs can run Qwen 1 8B?
52 consumer GPUs can run Qwen 1 8B 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 Qwen 1 8B?
163 devices with unified memory can run Qwen 1 8B at Q4_K_M (1.2 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.