Skip to content

Actions: google/tunix

Actions

Auto-assign issues and PRs

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
issue-auto-assign.yaml
133 workflow runs
133 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Fix: Update Qwen3 vLLM-JAX weight mapping Auto-assign issues and PRs #133: Pull request #1227 opened by xianglon-commits
10s
10s
[Tunix] Switch to safetensor API based loader when Pathways is enabled. Auto-assign issues and PRs #132: Pull request #1226 opened by copybara-service bot
39s
39s
Refactor reward computation structure Auto-assign issues and PRs #131: Pull request #1225 opened by copybara-service bot
9s
9s
Add eval datasets to tunix cli for grpo_main Auto-assign issues and PRs #130: Issue #1224 opened by andytwigg
12s
12s
[Tunix] batches log writes for efficiency and includes signal handlers to ensure logs are flushed on SIGINT/SIGTERM. Auto-assign issues and PRs #129: Pull request #1223 opened by copybara-service bot
12s
12s
Add sglang vllm config and update rollout engine setup in train_deepswe_nb.py. Auto-assign issues and PRs #128: Pull request #1222 opened by copybara-service bot
11s
11s
add support for GCS endpoint for perf trace writing Auto-assign issues and PRs #127: Pull request #1221 opened by copybara-service bot
9s
9s
Update issue auto-assignment script to async and clean up logic. Auto-assign issues and PRs #126: Pull request #1220 opened by copybara-service bot
10s
10s
Enable tied embedding for Qwen3-0.6B and Qwen3-1.7B. Auto-assign issues and PRs #125: Pull request #1219 opened by copybara-service bot
11s
11s
use unified trace_dir for the trace writer Auto-assign issues and PRs #124: Pull request #1218 opened by copybara-service bot
13s
13s
add qwen3 grpo example script with simplemath rewards Auto-assign issues and PRs #123: Pull request #1217 opened by andytwigg
11s
11s
add entroy loss, grad_norm to metrics Auto-assign issues and PRs #122: Pull request #1216 opened by copybara-service bot
10s
10s
Fix typos in README Auto-assign issues and PRs #121: Pull request #1215 opened by copybara-service bot
8s
8s
add simple_math reward_fn Auto-assign issues and PRs #120: Pull request #1214 opened by andytwigg
14s
14s
Add retries to Hugging Face Hub calls in tunix/oss/utils.py to mitigate gateway errors. Auto-assign issues and PRs #119: Pull request #1213 opened by copybara-service bot
11s
11s
[Tunix] Make returning logprobs from vLLM sampler configurable. Auto-assign issues and PRs #118: Pull request #1212 opened by copybara-service bot
11s
11s
Add RLOO advantage estimator to Tunix. Auto-assign issues and PRs #117: Pull request #1211 opened by copybara-service bot
13s
13s
force loss computation to be in fp32 Auto-assign issues and PRs #116: Pull request #1210 opened by copybara-service bot
8s
8s
add docs for perf tracing Auto-assign issues and PRs #115: Pull request #1209 opened by copybara-service bot
10s
10s
Add trajectory status to track limit. Auto-assign issues and PRs #114: Pull request #1208 opened by copybara-service bot
10s
10s
[BUG] remove flax==0.12.4 constraint for sglang Auto-assign issues and PRs #113: Issue #1207 opened by aolemila
10s
10s
fix flax==0.12.4 in tpu-tests.yml temporarily Auto-assign issues and PRs #112: Pull request #1206 opened by aolemila
9s
9s
[BUG] fail to run sglang_jax_lora_test.py with flax==0.12.5 Auto-assign issues and PRs #111: Issue #1205 opened by aolemila
9s
9s
add support for qwen3-base variants Auto-assign issues and PRs #110: Pull request #1204 opened by andytwigg
11s
11s
add pg_clipfrac to grpo_learner Auto-assign issues and PRs #109: Pull request #1203 opened by andytwigg
18s
18s