Spaces:
Paused
Paused
Commit History
26 days ago
-
Add Isaac Sim RTX renders of finished runs
Chandra KiranUnverified signaturePOST /simulate/run/{id}/isaac replays a run in Isaac Sim: worker/usd_scene.py builds a USD stage from the MuJoCo model with every body's pose time-sampled from the run (and a camera that tracks the robot like the MuJoCo video), and Isaac Sim only renders it frame by frame with RTX - no second physics engine, so both videos show the same motion. Deployed as its own Modal app on an L4 (worker/modal_isaac.py) so its large image never slows the main deploy; limited to ISAAC_DAILY_LIMIT renders a day (default 10). The USD stage is tested against MuJoCo's geom positions frame by frame; the Isaac Sim step itself needs the NGC image and a GPU and has not been run yet.2fc178d -
Track every backend job with its cost and stop at a monthly budget
Chandra KiranUnverified signatureRuns become one kind of job (sim, sweep, ros_check, review, export, isaac) in the same table, with the hardware it ran on and an estimated cost. New jobs are refused with 402 once the month's estimate reaches MONTHLY_CREDIT_LIMIT_USD (default $28). Sandboxed workers can return files inline; the API stores them in R2. GET /jobs and /jobs/{id} show any job, /report/{id} renders it with its follow-up jobs.1a36c6b -
Use GLM-5.3 via OpenRouter and run simulations on Modal (RunPod optional)
Chandra KiranUnverified signature- LLM settings are provider-neutral (LLM_API_KEY, LLM_BASE_URL, LLM_MODEL), defaulting to z-ai/glm-5.3 on OpenRouter; OpenCode gets the same endpoint as its only provider. - SIM_BACKEND selects modal (default) or runpod. worker/modal_app.py deploys the worker on a T4 (or CPU with MODAL_GPU=none) with a container cap and time limit. - Runs record backend + job_id; GPU time falls back to the worker's own timing on Modal. Existing run tables are migrated.da839c2