Edge0 120B Q3k — Hardware Requirements & GPU Compatibility
ChatFunctions
Edge0 120B Q3k is a 120B-parameter open language model from Edge0. At BF16 it needs about 264.00 GB of VRAM — see which GPUs and Macs can run it below.
Edge0 120B Q3k runs on a 512 GB+ Mac or a datacenter GPU
Best local pick → Mac Studio (M3 Ultra, 512GB) · BF16 · 269.5 GB · ~2 tok/s
Macs from 512 GB (BF16 · ~2 tok/s)
Other
Not set up for it?
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Edge0 120B Q3k
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 Edge0/Edge0-120B-q3k --max-model-len 16384 --tensor-parallel-size 8vLLM repo with its own quantisation
Download: Weights on Hugging Face
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on B300 (288 GB) (18 GB headroom) · BF16
- Generation speed
- ~20 tok/s
- generation speed
- Cost per 1M output tokens
- $97.86
- per 1M output tokens
How Much VRAM Does Edge0 120B Q3k Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | Runs on | |
|---|---|---|---|---|
| 264 GB | 275.7 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Runs on Instinct MI350X 288 GB |
Selected BF16
Brain floating point 16269.5GBat 16K context
- Weights
- 262.9 GB
- KV cache
- 6.3 GB
- Overhead
- 0.3 GB
Server-class: ~269.5 GB at 16K context · one Instinct MI350X 288 GB
Or .
Which Hardware Can Run Edge0 120B Q3k?
BF16 · 264 GBEdge0 120B Q3k (BF16) requires 264 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 344+ GB is recommended. No consumer GPU has enough memory. 1 device with unified memory can run Edge0 120B Q3k, including Mac Studio (M3 Ultra, 512GB).
Needs server-class hardware
Edge0 120B Q3k runs on a 512 GB+ Mac or a datacenter GPU · BF16 needs ~269.5 GB
| Hardware | Quant | Memory | Speed | Context | Hardware page | |
|---|---|---|---|---|---|---|
| Mac 512 GB · Mac Studio (M3 Ultra, 512GB) | BF16 | 269.5 GB | ~2 tok/s | 32K ctx | Mac Studio → | BF16 · 269.5 GB · ~2 tok/s · 32K ctx |
| Instinct MI350X 288 GB · one GPU | BF16 | 269.5 GB | ~18 tok/s | 32K ctx | Instinct MI350X → | BF16 · 269.5 GB · ~18 tok/s · 32K ctx |
| 4× H100 SXM 80 GB · 320 GB | BF16 | 269.5 GB | — | 32K ctx | H100 SXM → | BF16 · 269.5 GB · 32K ctx |
| 3× H200 SXM 141 GB · 423 GB | BF16 | 269.5 GB | — | 32K ctx | H200 SXM → | BF16 · 269.5 GB · 32K ctx |
| 2× B200 192 GB · 384 GB | BF16 | 269.5 GB | — | 32K ctx | B200 → | BF16 · 269.5 GB · 32K ctx |
Also: Instinct MI350X · Instinct MI355X · B300 · Mac Studio (M3 Ultra, 512GB) · NVIDIA DGX A100 640GB · NVIDIA DGX H100
By VRAM tier · typical GPUs and Macs at BF16
Consumer GPUs (up to 32 GB) and Macs up to 256 GB: too small for this model (needs ≥ 269.5 GB)8 GB tier →12 GB tier →16 GB tier →24 GB tier →Macs →
Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| GPT OSS 120BOpenAI | 97.5 | 70.8 GB | —needs > 24 GB | same size |
| Ornith 1.5 122Bornith-ai | — | 75.5 GB | —needs > 24 GB | same size |
| NVIDIA Nemotron 3.5 Super VL 120B A12B BF16NVIDIA | — | 87.7 GB | —needs > 24 GB | same size |
| Ling 3.0 Flash VLInclusion AI | — | 76 GB | —needs > 24 GB | same size |
| Leanstral 1.5 119B A6BMistral AI | — | 84 GB | —needs > 24 GB | same size |
| Qwen1.5 110BAlibaba | — | 72.4 GB | —needs > 24 GB | same size |
Frequently Asked Questions
How much VRAM does Edge0 120B Q3k need?
Edge0 120B Q3k requires 264 GB of VRAM at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 120B × 16 bits ÷ 8 = 240 GB
KV Cache + Overhead ≈ 24 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
Can NVIDIA GeForce RTX 5090 run Edge0 120B Q3k?
No — Edge0 120B Q3k requires at least 264 GB at BF16, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
Can I run Edge0 120B Q3k on a Mac?
Yes — Mac Studio (M3 Ultra, 512GB) and above can run Edge0 120B Q3k. Apple Silicon uses unified memory, so the model shares RAM with the system. At BF16 you need at least 264 GB of usable unified memory (RAM minus macOS overhead).
Can I run Edge0 120B Q3k locally?
Yes — Edge0 120B Q3k can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Edge0 120B Q3k?
At BF16, Edge0 120B Q3k can reach ~18 tok/s on AMD Instinct MI350X. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B300 → 8000 ÷ 264.0 × 0.65 = ~20 tok/s
Estimated speed at BF16 (264 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Edge0 120B Q3k?
At BF16, the download is about 240.00 GB.
Which GPUs can run Edge0 120B Q3k?
No single consumer GPU has enough VRAM to run Edge0 120B Q3k at BF16 (264 GB). Multi-GPU or professional hardware is required.
Which devices can run Edge0 120B Q3k?
1 device with unified memory can run Edge0 120B Q3k at BF16 (264 GB), including Mac Studio (M3 Ultra, 512GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.