LFM2.5 24B A2B ONNX — Hardware Requirements & GPU Compatibility
ChatFunctions
LFM2.5 24B A2B ONNX is a 24B-parameter open language model from Liquid AI in the LFM2.5 family. It supports a context window of up to 131,072 tokens. At Q4_K_M it needs about 14.87 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 12 GB (Q2_K · 64K)Best on 24 GB (Q4_K_M · ~220 tok/s · 128K)Macs from 32 GB (~38 tok/s)
Other
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run LFM2.5 24B A2B ONNX
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
vllm serve LiquidAI/LFM2.5-24B-A2B-ONNX --max-model-len 16384vLLM repo with its own quantisation
Download: Weights on Hugging Face
Other ways
each opens step-by-stepRun in cloud
Fits on RTX A5000 (24 GB) (9 GB headroom) · Q4_K_M
- Generation speed
- ~189 tok/s
- generation speed
- Cost per 1M output tokens
- $0.23
- per 1M output tokens
How Much VRAM Does LFM2.5 24B A2B ONNX Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 10.7 GB | 11.2 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 128K context · ~257 tok/s | |
| 12.2 GB | 12.7 GB | 3-bit medium — 3-bit medium quantization | Fits: 128K context · ~243 tok/s | |
| 14.9 GB | 15.4 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 128K context · ~220 tok/s | |
| 17.6 GB | 18.1 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 128K context · ~201 tok/s | |
| 20.3 GB | 20.8 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 128K context · ~185 tok/s | |
| 24.5 GB | 25 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Does not fit: needs 24.6 GB | |
| 48.5 GB | 49 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Does not fit: needs 48.6 GB |
Selected Q4_K_M
4-bit medium quantization15.0GBat 16K context
- Weights
- 14.4 GB
- KV cache
- 0.3 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 9.0 GB free · ~220 tok/s
≥50 tok/s needs ≥133 GB/s · RTX 5090 ~284, RTX 3090 Ti ~220
Which Hardware Can Run LFM2.5 24B A2B ONNX?
Q4_K_M · 14.9 GBLFM2.5 24B A2B ONNX (Q4_K_M) requires 14.9 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 20+ GB is recommended. Using the full 131K context window can add up to 2.5 GB, bringing total usage to 17.4 GB. 26 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 5080. 125 devices with unified memory can run LFM2.5 24B A2B ONNX, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Mini M5 Pro (24 GB).
16–18 GB
16–18 GB tier →fits · 1.0 GB free · up to 32K ctx
18 GPUs
RTX 5080~214Simulate output on RTX 5080RTX 5070 Ti~207Simulate output on RTX 5070 TiRTX 4080 SUPER~185Simulate output on RTX 4080 SUPER20 GB
fits · 5.0 GB free · up to 128K ctx
1 GPU
RX 7900 XT~89Simulate output on RX 7900 XT32–40 GB
fits · 17.0 GB free · up to 128K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~284Simulate output on RTX 5090M5 Max36 GB~78Simulate output on Mac Studio M5 Max (36 GB)M532 GB~45Simulate output on MacBook Air 13" M5 (32 GB)Orin~73Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~46Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Alternatives to LFM2.5 24B A2B ONNX
LFM2.5 family · 14 models →Similar models
4| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Gemma 4 26B A4B ITGoogle | 105.4 | 16.5 GB | ~171 | same size |
| Magistral Small 2509Mistral AI | 85.5 | 18.1 GB | ~43 | same size |
| Ariarhymes-ai | — | 20.2 GB | ~168 | same size |
| EuroLLM 22B Instruct 2512utter-project | — | 17.5 GB | ~46 | same size |
Specifications
- Publisher
- Liquid AI
- Family
- LFM2.5
- Parameters
- 24B
- Active parameters
- 2.5B per token
- Architecture
- Lfm2MoeForCausalLM
- Context Length
- 131,072 tokens
- Vocabulary Size
- 128,000
- Release Date
- 2026-10-09
- License
- Other
Frequently Asked Questions
How much VRAM does LFM2.5 24B A2B ONNX need?
LFM2.5 24B A2B ONNX requires 14.9 GB of VRAM at Q4_K_M, or 48.5 GB at BF16. Full 131K context adds up to 2.5 GB (17.4 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 24B × 4.8 bits ÷ 8 = 14.4 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 ≈ 3 GB (at full 131K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 4090 run LFM2.5 24B A2B ONNX?
Yes, at Q6_K (20.3 GB) or lower. Higher quantizations like Q8_0 (24.5 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
What's the best quantization for LFM2.5 24B A2B ONNX?
For LFM2.5 24B A2B ONNX, Q4_K_M (14.9 GB) offers the best balance of quality and VRAM usage. Q5_K_M (17.6 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 10.7 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run LFM2.5 24B A2B ONNX on a Mac?
Yes, but only at lower quantizations. The smallest Mac that can run LFM2.5 24B A2B ONNX is Mac Mini M1 (16 GB) at Q2_K; 135 of the 156 Macs we list can run it at some quantization. For Q4_K_M (14.9 GB) you need a Mac with more unified memory.
Can I run LFM2.5 24B A2B ONNX locally?
Yes — LFM2.5 24B A2B ONNX can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is LFM2.5 24B A2B ONNX?
At Q4_K_M, LFM2.5 24B A2B ONNX can reach ~120 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~220 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 → 1.5 GB active ÷ (8000 × 0.65) = 0.296 ms, plus 40 layers × 0.055 ms = 2.200 ms, so 1000 ÷ 2.496 ms = ~401 tok/s
Estimated speed at Q4_K_M (14.9 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of LFM2.5 24B A2B ONNX?
At Q4_K_M, the download is about 14.40 GB. The full-precision BF16 version is 48.00 GB. The smallest option (Q2_K) is 10.20 GB.
Which GPUs can run LFM2.5 24B A2B ONNX?
26 consumer GPUs can run LFM2.5 24B A2B ONNX at Q4_K_M. Top options include AMD Radeon RX 7900 XTX, NVIDIA GeForce RTX 3090, NVIDIA GeForce RTX 3090 Ti, AMD Radeon RX 6800. 7 GPUs have plenty of headroom for comfortable inference.
Which devices can run LFM2.5 24B A2B ONNX?
125 devices with unified memory can run LFM2.5 24B A2B ONNX at Q4_K_M (14.9 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB), Beelink GTR9 Pro (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.