AI Squawk
the real-time terminal for AI news
feed
·
changes
·
pricing
·
deprecations
·
check
·
api
·
subscribe
squawk
↗
← all sources
RSS for HN "DeepSeek" (>=50pt)
21h ago
TOOLING
DeepSeek Ultrafast
#
HN "DeepSeek" (>=50pt)
6d ago
TOOLING
Llama-modchip: 284B DeepSeek on 64GB RAM, no GPU
#
🐋 Run 284B DeepSeek-V4-Flash (<59GiB RAM, 44.4 tok/s pp, 9.1-14.7 tok/s) & 30B MoEs (25-111 tok/s) on a pure 64GB CPU at 256K ctx…
—
HN "DeepSeek" (>=50pt)