Search results

[RFC] Proposal for a central transfer planner between peers

Status: Open.

#293 In ai-dynamo/modelexpress;

· wseaton opened on May 20, 2026

vLLM DP workers can collide on MX P2P serving ports when CUDA device is unindexed

Status: Open.

#266 In ai-dynamo/modelexpress;

· zhengluo-nv opened on May 6, 2026

[Bug] Lint PR workflow permission for fork PRs

Status: Open.

#263 In ai-dynamo/modelexpress;

· emricksini-h opened on May 5, 2026

Generalize vLLM Cold Start Loading

Status: Open.

#253 In ai-dynamo/modelexpress;

· noa-neria opened on Apr 26, 2026

Fix support for MS distributed streaming in ModelStreamerStrategy

fix

Status: Open.

#240 In ai-dynamo/modelexpress;

· noa-neria opened on Apr 15, 2026

[Feature] Optimize model downloads by skipping duplicate weight formats

Status: Open.

#173 In ai-dynamo/modelexpress;

· Jont828 opened on Mar 20, 2026

[Feature] Add support for NIXL backend selection for p2p transfer

Status: Open.

#155 In ai-dynamo/modelexpress;

· erezzarum opened on Mar 4, 2026

[Feature] Download acceleration: multi-file parallelism, retry resilience, and tunable download parameters

Status: Open.

#149 In ai-dynamo/modelexpress;

· baladithyab opened on Feb 25, 2026

[Feature] Distributed Downloading

Status: Open.

#78 In ai-dynamo/modelexpress;

· tlipoca9 opened on Sep 24, 2025