Models / LFM2.5 8B A1B

LFM2.5 8B A1B

MoE consumer
HuggingFace → HuggingBay → Liquid AI → Find for my hardware

Hybrid MoE, 1B active params. Absurdly fast for tight-budget scenarios. Best for latency-critical edge use cases.

chat general
Parameters
8.0B
Context
32k
License
apache 2.0
Developer
Liquid AI
Origin
🇺🇸 USA
Released
Sep 2025

Scores

Coding
45
Reasoning
50
General
58

Related models

Kimi K3
2800.0B 104.0B active premier 610GB min RAM $0.68/M in Jul 16, 2026
Kimi K2.7 Code
1000.0B 32.0B active premier no local build $0.66/M in Jun 16, 2026
Kimi K2
1000.0B premier no local build $0.57/M in Jul 11, 2025
Command R+ 104B
104.0B enthusiast no local build $2.50/M in Apr 4, 2024

Save your hardware and every model page answers the real question: will it run on your machine, and how fast?

Join free - save your rig →

Run it locally

Per-quant memory needs and a static "can you run it?" reference - no rig entry required

Q4_K_M
2.5GB 2.5GB min 3.0GB rec
Balanced - the usual local sweet spot

The reference hardware

Schematic of the 4x H100 80GB (320GB) reference rig - 320GB VRAM, 13400 GB/s aggregate bandwidth
Schematic of the NVIDIA DGX Station 748GB reference rig - 748GB unified memory, 8000 GB/s aggregate bandwidth
Schematic of the 8x RTX 3090 rack (192GB) reference rig - 192GB VRAM, 7489 GB/s aggregate bandwidth
Schematic of the 4x RTX 5090 (128GB) reference rig - 128GB VRAM, 7168 GB/s aggregate bandwidth
Schematic of the AMD Instinct MI300X (192GB) reference rig - 192GB VRAM, 5324 GB/s aggregate bandwidth
Schematic of the 4x RTX 4090 (96GB) reference rig - 96GB VRAM, 4032 GB/s aggregate bandwidth
Schematic of the 2x RTX 5090 (64GB) reference rig - 64GB VRAM, 3584 GB/s aggregate bandwidth
Schematic of the 2x RTX 3090 (48GB) reference rig - 48GB VRAM, 1872 GB/s aggregate bandwidth
Schematic of the Single RTX 5090 (32GB) reference rig - 32GB VRAM, 1792 GB/s aggregate bandwidth
Schematic of the RTX PRO 6000 Blackwell (96GB) reference rig - 96GB VRAM, 1792 GB/s aggregate bandwidth
Schematic of the Mac Studio M4 Ultra 192GB reference rig - 192GB unified memory, 1092 GB/s aggregate bandwidth
Schematic of the Mac Studio M4 Ultra 512GB reference rig - 512GB unified memory, 1092 GB/s aggregate bandwidth
Schematic of the Single RTX 4090 (24GB) reference rig - 24GB VRAM, 1008 GB/s aggregate bandwidth
Schematic of the MacBook Pro M5 Max 128GB reference rig - 128GB unified memory, 614 GB/s aggregate bandwidth
Schematic of the Single GTX 1080 Ti (11GB) reference rig - 11GB VRAM, 484 GB/s aggregate bandwidth
Schematic of the Dual EPYC 9004 + 768GB DDR5-4800 reference rig - 768GB unified memory, 460 GB/s aggregate bandwidth
Schematic of the DGX Spark 128GB unified reference rig - 128GB unified memory, 273 GB/s aggregate bandwidth
Schematic of the Ryzen AI Max+ 395 128GB reference rig - 128GB unified memory, 256 GB/s aggregate bandwidth
Schematic of the Jetson AGX Orin 64GB reference rig - 64GB unified memory, 204 GB/s aggregate bandwidth
Schematic of the Epyc + 512GB DDR4-3200 + 2x RTX 3090 reference rig - 560GB unified memory, 204 GB/s aggregate bandwidth
Schematic of the Epyc + 512GB DDR4-2400 + 2x RTX 3090 reference rig - 560GB unified memory, 153 GB/s aggregate bandwidth
Schematic of the NVIDIA Jetson Orin NX 16GB reference rig - 16GB unified memory, 102 GB/s aggregate bandwidth

22 reference configs, drawn in-house. Scroll for more.

Can you run it? - reference rigs

Rig Q4_K_M
4x H100 80GB (320GB) fast 23523.8t/s
NVIDIA DGX Station 748GB fast 14044.1t/s
8x RTX 3090 rack (192GB) fast 13148.0t/s
4x RTX 5090 (128GB) fast 12583.5t/s
AMD Instinct MI300X (192GB) fast 9347.7t/s
4x RTX 4090 (96GB) fast 7078.2t/s
2x RTX 5090 (64GB) fast 6291.7t/s
2x RTX 3090 (48GB) fast 3287.0t/s
Single RTX 5090 (32GB) fast 3145.9t/s
RTX PRO 6000 Blackwell (96GB) fast 3145.9t/s
Mac Studio M4 Ultra 192GB fast 2091.3t/s
Mac Studio M4 Ultra 512GB fast 2091.3t/s
Single RTX 4090 (24GB) fast 1769.6t/s
MacBook Pro M5 Max 128GB fast 1175.9t/s
Single GTX 1080 Ti (11GB) fast 849.7t/s
Dual EPYC 9004 + 768GB DDR5-4800 fast 808.9t/s
DGX Spark 128GB unified fast 479.3t/s
Ryzen AI Max+ 395 128GB fast 449.4t/s
Jetson AGX Orin 64GB fast 359.5t/s
Epyc + 512GB DDR4-3200 + 2x RTX 3090 fast 359.5t/s
Epyc + 512GB DDR4-2400 + 2x RTX 3090 fast 269.7t/s
NVIDIA Jetson Orin NX 16GB fast 179.8t/s

Fit tiers use the same will-it-run logic as the rig finder. For comfortable fits, the badge reflects decode speed: fast >=20 t/s, ok 8-20 t/s, slow <8 t/s. t/s is a bandwidth estimate, not a measured benchmark.

Download options

Q4_K_M official -5% vs fp16
2.5GB dl 2.5GB min 3.0GB rec
REC RAM vs largest quant

Or run it in the cloud

No per-token API provider pricing tracked for LFM2.5 8B A1B yet. For flagship list prices, see the calculator.

PRICE HISTORY

Inference cost over time

Data accumulates from the first daily sync - longer ranges populate over time. Prices come from OpenRouter snapshots, not a historical API.

Loading price history...