Lumma 0.6B Instruct — Hardware Requirements & GPU Compatibility
Chat
Lumma 0.6B Instruct is a 649M-parameter open language model from FrontiersMind. It supports a context window of up to 12,288 tokens. At BF16 it needs about 1.78 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 8 GB (BF16 · 12K)Best on 24 GB (BF16 · ~368 tok/s · 12K)Macs from 8 GB (BF16 · ~40 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Lumma 0.6B Instruct
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
vllm serve FrontiersMind/Lumma-0.6B-Instruct --max-model-len 12288vLLM repo with its own quantisation
Download: Weights on Hugging Face
Other ways
each opens step-by-stepRun in cloud
Fits on RTX 3060 12GB (9 GB headroom) · BF16
- Generation speed
- ~132 tok/s
- generation speed
- Cost per 1M output tokens
- $0.15
- per 1M output tokens
How Much VRAM Does Lumma 0.6B Instruct Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 1.8 GB | — | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 12K context · ~368 tok/s |
Selected BF16
Brain floating point 162.7GBat 12K context
- Weights
- 1.3 GB
- KV cache
- 1.1 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 21.3 GB free · ~368 tok/s
≥50 tok/s needs ≥137 GB/s · RTX 5090 ~654, RTX 3090 Ti ~368
Which Hardware Can Run Lumma 0.6B Instruct?
BF16 · 1.8 GBLumma 0.6B Instruct (BF16) requires 1.8 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 3+ GB is recommended. Using the full 12K context window can add up to 0.9 GB, bringing total usage to 2.7 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 163 devices with unified memory can run Lumma 0.6B Instruct, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 5.3 GB free · up to 12K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~278Simulate output on RTX 3080RTX 2080 Ti~225Simulate output on RTX 2080 TiRTX 3070 Ti~222Simulate output on RTX 3070 TiM3*8 GB~40Simulate output on iMac 24" M3 (8 GB)Orin Nano~37Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~27Simulate output on iPhone 1720 GB
fits · 17.3 GB free · up to 12K ctx
1 GPU
RX 7900 XT~270Simulate output on RX 7900 XT32–40 GB
fits · 29.3 GB free · up to 12K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~654Simulate output on RTX 5090M5 Max36 GB~181Simulate output on Mac Studio M5 Max (36 GB)M532 GB~60Simulate output on MacBook Air 13" M5 (32 GB)Orin~75Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~49Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at BF16
Similar models
1| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Qwen1.5 0.5B ChatAlibaba | — | 2.3 GB | ~753 | same size |
Specifications
- Publisher
- FrontiersMind
- Parameters
- 649M
- Architecture
- LummaForCausalLM
- Context Length
- 12,288 tokens
- Vocabulary Size
- 131,072
- Release Date
- 2026-08-18
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does Lumma 0.6B Instruct need?
Lumma 0.6B Instruct requires 1.8 GB of VRAM at BF16. Full 12K context adds up to 0.9 GB (2.7 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 649M × 16 bits ÷ 8 = 1.3 GB
KV Cache + Overhead ≈ 0.5 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 ≈ 1.4 GB (at full 12K context)
VRAM usage by quantization
Can I run Lumma 0.6B Instruct on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run Lumma 0.6B Instruct. Apple Silicon uses unified memory, so the model shares RAM with the system. At BF16 you need at least 1.8 GB of usable unified memory (RAM minus macOS overhead).
Can I run Lumma 0.6B Instruct locally?
Yes — Lumma 0.6B Instruct can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Lumma 0.6B Instruct?
At BF16, Lumma 0.6B Instruct can reach ~2697 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~368 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 ÷ 1.78 × 0.65 = ~2921 tok/s
Estimated speed at BF16 (1.8 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Lumma 0.6B Instruct?
At BF16, the download is about 1.30 GB.
Which GPUs can run Lumma 0.6B Instruct?
52 consumer GPUs can run Lumma 0.6B Instruct at BF16. 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 Lumma 0.6B Instruct?
163 devices with unified memory can run Lumma 0.6B Instruct at BF16 (1.8 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.