Understand this story
On October 22, 2026, LMSYS published a blog post describing its use of Miles, SGLang, Megatron-LM, and Harbor to conduct fully asynchronous agentic reinforcement learning on Qwen3-30B-A3B-Instruct on AMD Instinct MI355X. On a 100-task stratified subset of SWE-bench Verified, pass@1 improved from 7.3% to 27.5%. Training was conducted in two stages: first SFT on approximately 4,000 public Open-SWE-Traces trajectories, then GRPO on approximately 1,350 SWE-smith tasks. After fixing the model and RL recipe, scaling inference from 1 node to 3 nodes with 8 GPUs each, switching to E2B cloud sandboxes, and removing the concurrency limit raised rollout throughput from about 860 to about 3,480 trajectories per hour; on the same 8-GPU node, limiting trajectory length reduced single-step training time from 850 seconds to 284 seconds.
AI-generated from public reports
Latest development · Report
LMSYS uses Miles to train a SWE agent on MI355X, SWE-bench Verified…Developments
1 confirmed developments
- Source:LMSYS Blog
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Oct 22
LMSYS BlogFirst-party
LMSYS uses Miles to train a SWE agent on MI355X, SWE-bench Verified…
LMSYS used Miles, SGLang, Megatron-LM, and Harbor to perform fully asynchronous agent reinforcement learning on Qwen3-30B-A3B-Instruct on AMD Instinct MI355X, raising pass@1 from 7.3% to 27.5% on a 100-task stratified subset of SWE-bench Verified. Training was conducted in two stages: first SFT on approximately 4,000 public trajectories from Open-SWE-Traces, then GRPO on approximately 1,350 SWE-smith tasks. After fixing the model and RL recipe, scaling inference from 1 node to 3 8-GPU nodes, switching to E2B cloud sandboxes, and removing the concurrency limit increased rollout throughput from approximately 860 to approximately 3,480 trajectories/hour; limiting trajectory length on the same 8-GPU node reduced single-step training time from 850 seconds to 284 seconds.