LoreonLabsPlatform

Overview

Intelligence

Markets
Builders
Research
Narratives
Ecosystems
Launchpads

Discover

Search
Sources

Python

verl

verl: Volcano Engine Reinforcement Learning for LLMs

PythonEmerging

Stars

—

Forks

—

Contributors

8

Last push

5mo ago

Recent commits

Latest commits.

[megatron] feat: Using MTP in RL Training and Inference (#4936)
df340d7arron5mo ago
[sglang, rollout] feat: support sglang as rollout engine in fully async policy (#4191)
f104dfaPeng Zhang5mo ago
[doc, data] fix: resolve broken documentation hyperlinks (#4970)
5053d42aphrodite10285mo ago
[training_utils] A bug that caused device selection in group statistics to fail has been covered by tests. (#4967)
75d8b00JohnConnor1235mo ago
[training_utils] fix: Correct Attention TFLOPS estimation & fix CI (#4959)
54582f1HaochenYuan

5mo ago

[trainer] fix: pass scores device type to `group_mean_std` call (#4962)

5689fd7ℍ𝕠𝕝𝕝𝕠𝕨 𝕄𝕒𝕟5mo ago

[training_utils] fix: correctly `_resolve_device` when not specified (#4961)

e9c43b9ℍ𝕠𝕝𝕝𝕠𝕨 𝕄𝕒𝕟5mo ago

[env] fix: upgrade torch, cudnn and deps versions in vllm image to fix performance issue (#4960)

65eb5a1Begunner5mo ago

Top contributors

Builders behind this project.

eric-haibin-lin