Skip to content

Add thumbnail image to bitwise consistent RL blog post #114

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Nov 12, 2025
+1 −0
Merged

Add thumbnail image to bitwise consistent RL blog post #114

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
d23fd3b
Add thumbnail image to bitwise consistent RL blog post
mgoin Nov 12, 2025
4fd77f4
Change image for bitwise consistent RL blog post
mgoin Nov 12, 2025
a2bd092
Merge branch 'main' into bitwise-thumbnail
mgoin Nov 12, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions _posts/2025-11-10-bitwise-consistent-train-inference.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,7 @@
layout: post
title: "No More Train-Inference Mismatch: Bitwise Consistent On-Policy Reinforcement Learning with vLLM and TorchTitan"
author: "vLLM and TorchTitan Teams"
image: /assets/figures/2025-11-10-bitwise-exact-rl/reward-comparison.png
---

We demonstrate an open-source bitwise consistent on-policy RL run with [TorchTitan](https://github.com/pytorch/torchtitan) as the training engine and [vLLM](https://github.com/vllm-project/vllm) as the inference engine. Built on top of [vLLM's recent work on batch-invariant inference](https://docs.vllm.ai/en/latest/features/batch_invariance/), we show how to run an RL fine-tune of Qwen3 1.7B with bitwise matching training and inference numerics in [our open-sourced instructions](https://github.com/pytorch/torchtitan/tree/main/torchtitan/experiments/deterministic_vllm_rl):
Expand Down