willitrun·ai

Liquid AILiquid AI

LFM2.5 350M

huggingfaceHuggingFace
80.5KDownloads363LikesMar 2026Veröffentlicht128K TokenKontextOtherLizenz30 BasisQualität

LFM2.5 350M (0.3499999940395355B parameters) requires approximately 1.9 GB of VRAM with Q4_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 3 GB of VRAM.

Loslegen

— kopieren & einfügen, um lokal auszuführen

Copy-paste commands to run LFM2.5 350M on your machine.

Run

lms load LFM2.5-350M && lms server start

Quick specs

Parameters0.35B
Architecturedense
Context128K tokens
Modalitytext
Min RAM0.1 GB
Rec. RAM0.2 GB (Q4_K_M)
LicenseOther
FamilyLFM2

About this model

LFM2.5-350M is Liquid AI's ultra-compact 354M-parameter model for on-device and embedded deployment, using the hybrid convolution + attention LFM2 backbone with a 128K context window.

Verwandte Modelle

Deine Hardware

Erkennung...

Schnellauswahl

Intel
Bestes BudgetC
Intel Arc A380 6GB~$139 — 5 tok/s
NVIDIA
Beste GesamtwahlC
GTX 1650 4GB~$149 — 5 tok/s

Beste Hardware

Top-Empfehlungen für LFM2.5 350M

NVIDIA
RTX 3050 Ti Laptop 4GBB
4 GB
Intel
Intel Arc A370M 4GBC
4 GB
NVIDIA
GTX 1650 4GBC
4 GB
NVIDIA
RTX 4050 Laptop 6GBC
6 GB
NVIDIA
RTX 2060 6GBC
6 GB

Dieses Modell ausführen

LFM2.5 350M on RTX 3050 Ti Laptop 4GBLFM2.5 350M on Intel Arc A370M 4GBLFM2.5 350M on GTX 1650 4GB

Quantisierungsoptionen

VRAM-Schätzungen nach Quantisierungsstufe

No hardware detected — fit column shows raw VRAM estimates

QuantBitsVRAMQualityFit
Q2_K
2
0.1 GB
Low—
Q3_K_S
3
0.2 GB
Low—
NVFP4
4
0.2 GB
Medium—
Q4_K_M
4
0.2 GB
Medium—
Q5_K_M
5
0.3 GB
High—
Q6_K
6
0.3 GB
High—
Q8_0
8
0.4 GB
Very High—
F16
16
0.7 GB
Maximum—

Hardware-Kompatibilität

Eignungsschätzungen für alle Hardware

Rechner öffnen

Computing compatibility...

Speicheraufschlüsselung

Reference: RTX 2060 6GB

Weights0.2 GB
KV Cache0.2 GB
Runtime0.9 GB
Headroom0.6 GB

Häufig gestellte Fragen

FAQ — LFM2.5 350M

Siehe auch

QuantisierungsleitfadenBewertungsmethodikRechner öffnen