llmrun
Edge0BailingMoeV3ForCausalLM

Edge0 8B A1B Preview — Hardware Requirements & GPU Compatibility

Chat

Edge0 8B A1B Preview is a 7.9B-parameter open language model from Edge0. It supports a context window of up to 131,072 tokens. At BF16 it needs about 16.45 GB of VRAM — see which GPUs and Macs can run it below.

7.9B params · MoE1.4B active131K contextApache 2.05.1K downloads79 likes
S

Runs great on a 24 GB GPU

Runs on24 GB BF16 · ~173 tok/s · 128KMacs from24 GB BF16 · ~54 tok/s

Apache 2.0

Reference GPU · RTX 4090 24 GB

Estimates, ±20% · how we estimate

Run Edge0 8B A1B Preview

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
one command · needs vLLM installed
$vllm serve Edge0/Edge0-8B-A1B-preview --max-model-len 16384

vLLM repo with its own quantisation

Download: Weights on Hugging Face

install, OS, other quants

Ways to runOther ways

each opens step-by-step

Run in cloud

Runs on RTX 3090 (24 GB) (7 GB headroom) · BF16

Generation speed
~163 tok/s
generation speed
Cost per 1M output tokens
$0.34
per 1M output tokens
Deployon Vast.aiBest valuefrom $0.20/hr (opens in new tab)
Compare GPUs →
or

How Much VRAM Does Edge0 8B A1B Preview Need?

Select a quantization level; the listed hardware adheres to it.

QuantizationVRAM+32KOn RTX 4090 (reference)
16.4 GB17.1 GB16-bit · bfloat — Brain floating point 16 — preferred for trainingFits: 128K context · ~173 tok/s
est.calculated, no published file yet

Selected BF16

Brain floating point 16

16.6GBat 16K context

Weights
15.9 GB
KV cache
0.5 GB
Overhead
0.3 GB

Fits RTX 4090 (reference) · 7.4 GB free · ~173 tok/s

≥50 tok/s needs ≥241 GB/s · RTX 5090 ~261, RTX 3090 Ti ~173

Which Hardware Can Run Edge0 8B A1B Preview?

BF16 · 16.4 GB

Edge0 8B A1B Preview (BF16) requires 16.4 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 22+ GB is recommended. Using the full 131K context window can add up to 3.3 GB, bringing total usage to 19.8 GB. 8 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 122 devices with unified memory can run Edge0 8B A1B Preview, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Mini M5 Pro (24 GB).

By VRAM tier · typical GPUs and Macs at BF16
8–16 GB · RTX 4060, RTX 3060 12GB, RTX 5060 Ti 16GBDoesn't fit — smallest quant needs 16.6 GB8 GB tier →12 GB tier →16 GB tier →
24 GB · RTX 4090referenceBF16 · 16.6 GB · ~173 tok/s · 128K24 GB tier →
Mac 32 GB · Mac Mini M4BF16 · 16.6 GB · ~25 tok/s · 128KMacs →
Mac 64 GB · Mac Studio M4 MaxBF16 · 16.6 GB · ~80 tok/s · 128KMacs →

Alternatives to Edge0 8B A1B Preview

Similar models

6
ModelScoreVRAMtok/sWhy it is similar
Ling 3.0 TinyInclusion AI—5.5 GB~364same size
LFM2.5 8B A1BLiquid AI—5.6 GB~335same size
BitCPM CANN 8BOpenBMB—5.6 GB~127same size
Gemma 4 E4BGoogle—5.6 GB~123same size
Phi Mini MoE InstructMicrosoft—5.4 GB~239same size
Gemma 3n E4B ITGoogle—7.7 GB~127same size

Specifications

Publisher
Edge0
Parameters
7.9B
Active parameters
1.4B per token
Architecture
BailingMoeV3ForCausalLM
Context Length
131,072 tokens
Vocabulary Size
157,184
Release Date
2026-09-08
License
Apache 2.0

Resources

Frequently Asked Questions

How much VRAM does Edge0 8B A1B Preview need?

Edge0 8B A1B Preview requires 16.4 GB of VRAM at BF16. Full 131K context adds up to 3.3 GB (19.8 GB total).

VRAM = Weights + KV Cache + Overhead

Weights = 7.9B × 16 bits ÷ 8 = 15.8 GB

KV Cache + Overhead ≈ 0.6 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 ≈ 4 GB (at full 131K context)

VRAM usage by quantization

16.4 GB
19.8 GB

Learn more about VRAM estimation →

Can I run Edge0 8B A1B Preview on a Mac?

Yes — Mac Mini M5 Pro (24 GB) and 106 other Macs can run Edge0 8B A1B Preview. Apple Silicon uses unified memory, so the model shares RAM with the system. At BF16 you need at least 16.4 GB of usable unified memory (RAM minus macOS overhead).

Can I run Edge0 8B A1B Preview locally?

Yes — Edge0 8B A1B Preview can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.

How fast is Edge0 8B A1B Preview?

At BF16, Edge0 8B A1B Preview can reach ~185 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~173 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.

tok/s = 1000 ÷ (active GB ÷ (bandwidth GB/s × efficiency) × 1000 + layers × routing ms)

Mixture-of-Experts: just the active experts get read for each token, along with a fixed routing cost per layer.

Example: NVIDIA B200 → 2.9 GB active ÷ (8000 × 0.65) = 0.564 ms, plus 24 layers × 0.055 ms = 1.320 ms, so 1000 ÷ 1.884 ms = ~531 tok/s

Estimated speed at BF16 (16.4 GB)

~531 tok/s
~173 tok/s
~531 tok/s
~446 tok/s

Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.

Learn more about tok/s estimation →

What's the download size of Edge0 8B A1B Preview?

At BF16, the download is about 15.85 GB.

Which GPUs can run Edge0 8B A1B Preview?

8 consumer GPUs can run Edge0 8B A1B Preview at BF16. 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 Edge0 8B A1B Preview?

122 devices with unified memory can run Edge0 8B A1B Preview at BF16 (16.4 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.