LFM2 24B A2B — Hardware Requirements & GPU Compatibility
Chat
LFM2 24B A2B is a 23.8B-parameter open language model from Liquid AI in the LFM2 family. It supports a context window of up to 128,000 tokens. At BF16 it needs about 48.16 GB of VRAM — see which GPUs and Macs can run it below.
3.8K downloads
351 likesSpecifications
- Publisher
- Liquid AI
- Family
- LFM2
- Parameters
- 23.8B
- Architecture
- Lfm2MoeForCausalLM
- Context Length
- 128,000 tokens
- Vocabulary Size
- 65,536
- Release Date
- 2026-02-24
- License
- Other
Get Started
HuggingFace
How Much VRAM Does LFM2 24B A2B Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| BF16est. | 16.00 | 48.2 GB | 58.5 GB | 47.69 GB | Brain floating point 16 — preferred for training |
est.= calculated VRAM estimate; no published GGUF file found for that quantization yet. Other rows are verified against real community uploads.
Which GPUs Can Run LFM2 24B A2B?
BF16 · 48.2 GBLFM2 24B A2B (BF16) requires 48.2 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 63+ GB is recommended. Using the full 128K context window can add up to 10.3 GB, bringing total usage to 58.5 GB. No single GPU has enough memory — multi-GPU or cluster setups are needed.
Which Devices Can Run LFM2 24B A2B?
BF16 · 48.2 GB22 devices with unified memory can run LFM2 24B A2B, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Studio (M3 Ultra, 96GB).
Runs great
— Plenty of headroom
Where to Download LFM2 24B A2B
Community quantizations of this model — GGUF for llama.cpp, Ollama, and LM Studio, plus AWQ/MLX variants where available.
Related Models
Frequently Asked Questions
- How much VRAM does LFM2 24B A2B need?
LFM2 24B A2B requires 48.2 GB of VRAM at BF16. Full 128K context adds up to 10.3 GB (58.5 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 23.8B × 16 bits ÷ 8 = 47.7 GB
KV Cache + Overhead ≈ 0.5 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 10.8 GB (at full 128K context)
VRAM usage by quantization
BF1648.2 GBBF16 + full context58.5 GB- Can NVIDIA GeForce RTX 5090 run LFM2 24B A2B?
No — LFM2 24B A2B requires at least 48.2 GB at BF16, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
- Can I run LFM2 24B A2B on a Mac?
LFM2 24B A2B requires at least 48.2 GB at BF16, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run LFM2 24B A2B locally?
Yes — LFM2 24B A2B can run locally on consumer hardware. At BF16 quantization it needs 48.2 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is LFM2 24B A2B?
At BF16, LFM2 24B A2B can reach ~118 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 B200 → 8000 ÷ 48.2 × 0.65 = ~375 tok/s
Estimated speed at BF16 (48.2 GB)
~375 tok/s~375 tok/s~338 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of LFM2 24B A2B?
At BF16, the download is about 47.69 GB.
- Which GPUs can run LFM2 24B A2B?
No single consumer GPU has enough VRAM to run LFM2 24B A2B at BF16 (48.2 GB). Multi-GPU or professional hardware is required.
- Which devices can run LFM2 24B A2B?
23 devices with unified memory can run LFM2 24B A2B at BF16 (48.2 GB), including ASUS Ascent GX10, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB), 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.