LiquidAI releases QAD checkpoints, 4-bit models retain 97% BF16 accuracy
JosephJacks_ · x · 2026-08-20
LiquidAI has released updated 4-bit checkpoints for the LFM2.5 series, trained with Quantization-Aware Distillation (QAD).
- Performance: The checkpoints recover accuracy lost to 4-bit quantization, reaching roughly 97% of their BF16 averages while retaining low memory footprint and high throughput.
- Usage: A developer shared a command to run the 2.6B model via llama-server with specific parameters (131k context, temp 0.1, top-k 50), noting impressive performance on minimal hardware.
Related event: Liquid AI Releases QAD 4-bit Models Retaining 97% Performance(2 posts)→
More from Infra
- Qwen3.8-27B Test: Lower KV Cache Quantization Impacts Reasoning Quality — fbms2 · 2026-08-20
- Trading speed for cost on DGX Spark, RTX 5090 suggested for speed boost — daniel_mac8 · 2026-08-20
- 1.5-Year Delay Cuts AI Data Center Value by 8.9%, Speed Key: Report — Beth_Kindig · 2026-08-20
- Clarification: OpenAI's 20% compute claim refers to monitoring overhead, not total capacity — sjgadler · 2026-08-20
- SkyPilot, VAST Data, and Partners Host AI Infra Meetup — skypilot_org · 2026-08-20
- PolymathicAI Releases The Well: A 15TB Collection of Physics Simulations — tom_doerr · 2026-08-20