LLaDA2.0 Mini — Hardware Requirements & GPU Compatibility
ChatFunctions
LLaDA2.0 Mini is a 16.3B-parameter open language model from Inclusion AI. It supports a context window of up to 32,768 tokens. At Q4_K_M it needs about 10.14 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 8 GB (Q2_K · 16K)Best on 24 GB (Q4_K_M · ~406 tok/s · 32K)Macs from 16 GB (Q4_K_M · ~96 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run LLaDA2.0 Mini
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 inclusionAI/LLaDA2.0-mini --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 A4000 (16 GB) (5 GB headroom) · Q4_K_M
- Generation speed
- ~240 tok/s
- generation speed
- Cost per 1M output tokens
- $0.10
- per 1M output tokens
How Much VRAM Does LLaDA2.0 Mini Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 7.3 GB | 8.6 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 32K context · ~481 tok/s | |
| 8.3 GB | 9.6 GB | 3-bit medium — 3-bit medium quantization | Fits: 32K context · ~451 tok/s | |
| 10.1 GB | 11.4 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 32K context · ~406 tok/s | |
| 12.0 GB | 13.2 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 32K context · ~369 tok/s | |
| 13.8 GB | 15.1 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 32K context · ~338 tok/s | |
| 16.6 GB | 17.9 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 32K context · ~300 tok/s | |
| 32.9 GB | 34.2 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Does not fit: needs 33.5 GB |
Selected Q4_K_M
4-bit medium quantization10.7GBat 16K context
- Weights
- 9.8 GB
- KV cache
- 0.7 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 13.3 GB free · ~406 tok/s
≥50 tok/s needs ≥73 GB/s · RTX 5090 ~536, RTX 3090 Ti ~406
Which Hardware Can Run LLaDA2.0 Mini?
Q4_K_M · 10.1 GBLLaDA2.0 Mini (Q4_K_M) requires 10.1 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 14+ GB is recommended. Using the full 33K context window can add up to 1.3 GB, bringing total usage to 11.4 GB. 38 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3080 Ti. 150 devices with unified memory can run LLaDA2.0 Mini, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Mini M2 Pro (16 GB).
12 GB
12 GB tier →fits · 1.3 GB free · up to 32K ctx
10 GPUs
RTX 3080 Ti~384Simulate output on RTX 3080 TiRTX 5070~318Simulate output on RTX 5070RTX 4070~261Simulate output on RTX 407020 GB
fits · 9.3 GB free · up to 32K ctx
1 GPU
RX 7900 XT~171Simulate output on RX 7900 XT32–40 GB
fits · 21.3 GB free · up to 32K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~536Simulate output on RTX 5090M5 Max36 GB~148Simulate output on Mac Studio M5 Max (36 GB)M532 GB~81Simulate output on MacBook Air 13" M5 (32 GB)Orin~128Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~82Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Similar models
4| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Kimi VL A3B InstructMoonshot AI | — | 10.7 GB | ~208 | same size |
| Moonlight 16B A3BMoonshot AI | — | 10.1 GB | ~229 | same size |
| DeepSeek v2 LiteDeepSeek | — | 10.2 GB | ~243 | same size |
| Deepseek Moe 16B ChatDeepSeek | — | 11.1 GB | ~231 | same size |
Specifications
- Publisher
- Inclusion AI
- Parameters
- 16.3B
- Active parameters
- 1.4B per token
- Architecture
- LLaDA2MoeModelLM
- Context Length
- 32,768 tokens
- Vocabulary Size
- 157,184
- Release Date
- 2025-11-25
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does LLaDA2.0 Mini need?
LLaDA2.0 Mini requires 10.1 GB of VRAM at Q4_K_M, or 32.9 GB at BF16. Full 33K context adds up to 1.3 GB (11.4 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 16.3B × 4.8 bits ÷ 8 = 9.8 GB
KV Cache + Overhead ≈ 0.3 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 ≈ 1.6 GB (at full 33K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 4090 run LLaDA2.0 Mini?
Yes, at Q8_0 (16.6 GB) or lower. Higher quantizations like BF16 (32.9 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
What's the best quantization for LLaDA2.0 Mini?
For LLaDA2.0 Mini, Q4_K_M (10.1 GB) offers the best balance of quality and VRAM usage. Q5_K_M (12.0 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 7.3 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run LLaDA2.0 Mini on a Mac?
Yes — Mac Mini M2 Pro (16 GB) and 133 other Macs can run LLaDA2.0 Mini. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 10.1 GB of usable unified memory (RAM minus macOS overhead).
Can I run LLaDA2.0 Mini locally?
Yes — LLaDA2.0 Mini can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is LLaDA2.0 Mini?
At Q4_K_M, LLaDA2.0 Mini can reach ~239 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~406 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 → 0.9 GB active ÷ (8000 × 0.65) = 0.172 ms, plus 20 layers × 0.055 ms = 1.100 ms, so 1000 ÷ 1.272 ms = ~786 tok/s
Estimated speed at Q4_K_M (10.1 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of LLaDA2.0 Mini?
At Q4_K_M, the download is about 9.75 GB. The full-precision BF16 version is 32.51 GB. The smallest option (Q2_K) is 6.91 GB.
Which GPUs can run LLaDA2.0 Mini?
38 consumer GPUs can run LLaDA2.0 Mini at Q4_K_M. Top options include AMD Radeon RX 6800, AMD Radeon RX 6800 XT, AMD Radeon RX 6900 XT, AMD Radeon RX 6700 XT. 26 GPUs have plenty of headroom for comfortable inference.
Which devices can run LLaDA2.0 Mini?
150 devices with unified memory can run LLaDA2.0 Mini at Q4_K_M (10.1 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.