Hugging Face
Models
Datasets
Spaces
Buckets
new
Docs
Enterprise
Pricing
Website
Tasks
HuggingChat
Collections
Languages
Organizations
Community
Blog
Posts
Daily Papers
Hardware
Learn
Discord
Forum
GitHub
Solutions
Team & Enterprise
Hugging Face PRO
Enterprise Support
Inference Providers
Inference Endpoints
Storage Buckets
Log In
Sign Up
ProprietaryLegal
/
dsv41-async-hostflag-handoff
Like
1
vllm
cuda
deadlock
custreamwaitvalue32
lazy-loading
dgx-spark
gb10
speculative-decoding
source-code
deepseek
deepseek-v4
rtx-3090
pli-labs
legal-ai
License:
apache-2.0
Model card
Files
Files and versions
xet
Community
Copy to bucket
new
main
dsv41-async-hostflag-handoff
/
code
/
dspark
/
spark
646 kB
Ctrl+K
Ctrl+K
1 contributor
History:
3 commits
ProprietaryLegal
PLI Labs: standalone docs, website links, newest results
33a54c8
verified
3 days ago
overlay
2026-10-23 update: measured on GB10 (out-of-process proxy: -0.9 to -3.9 %/round one card, -3.9 to -6.5 % two cards, TTFT parity, 0 watchdog fires); in-process cost finding; hardening (late-proxy corruption, start fallback, stall detection, respawn, shutdown) with independent-review evidence; full proxy code
3 days ago
remote_dspark
PLI Labs: standalone docs, website links, newest results
3 days ago
tests
PLI Labs: standalone docs, website links, newest results
3 days ago
tests_gpu
2026-10-23 update: measured on GB10 (out-of-process proxy: -0.9 to -3.9 %/round one card, -3.9 to -6.5 % two cards, TTFT parity, 0 watchdog fires); in-process cost finding; hardening (late-proxy corruption, start fallback, stall detection, respawn, shutdown) with independent-review evidence; full proxy code
3 days ago
tools
2026-10-23 update: measured on GB10 (out-of-process proxy: -0.9 to -3.9 %/round one card, -3.9 to -6.5 % two cards, TTFT parity, 0 watchdog fires); in-process cost finding; hardening (late-proxy corruption, start fallback, stall detection, respawn, shutdown) with independent-review evidence; full proxy code
3 days ago
ORDERING.md
Safe
14 kB
2026-10-23 update: measured on GB10 (out-of-process proxy: -0.9 to -3.9 %/round one card, -3.9 to -6.5 % two cards, TTFT parity, 0 watchdog fires); in-process cost finding; hardening (late-proxy corruption, start fallback, stall detection, respawn, shutdown) with independent-review evidence; full proxy code
3 days ago
OVERLAP.md
Safe
5.63 kB
2026-10-23 update: measured on GB10 (out-of-process proxy: -0.9 to -3.9 %/round one card, -3.9 to -6.5 % two cards, TTFT parity, 0 watchdog fires); in-process cost finding; hardening (late-proxy corruption, start fallback, stall detection, respawn, shutdown) with independent-review evidence; full proxy code
3 days ago
README.md
Safe
2.23 kB
2026-10-23 update: measured on GB10 (out-of-process proxy: -0.9 to -3.9 %/round one card, -3.9 to -6.5 % two cards, TTFT parity, 0 watchdog fires); in-process cost finding; hardening (late-proxy corruption, start fallback, stall detection, respawn, shutdown) with independent-review evidence; full proxy code
3 days ago
stage_overlay.sh
Safe
5.22 kB
2026-10-23 update: measured on GB10 (out-of-process proxy: -0.9 to -3.9 %/round one card, -3.9 to -6.5 % two cards, TTFT parity, 0 watchdog fires); in-process cost finding; hardening (late-proxy corruption, start fallback, stall detection, respawn, shutdown) with independent-review evidence; full proxy code
3 days ago
test_stage_overlay.sh
Safe
4.26 kB
2026-10-23 update: measured on GB10 (out-of-process proxy: -0.9 to -3.9 %/round one card, -3.9 to -6.5 % two cards, TTFT parity, 0 watchdog fires); in-process cost finding; hardening (late-proxy corruption, start fallback, stall detection, respawn, shutdown) with independent-review evidence; full proxy code
3 days ago