v0.11.0

0.11.0 (2026-07-07)

Features

• cli: clay variations command (08320fe)
• gpu-backend: thread seed through the contract (request→payload→runtime) (985999a)
• tools: generate_variations tool (shape reuse, GPU-gated) (2665aaf)
• variations: generate_variations — N seed-varied generations via the shape pipeline (c9a3f85)

Bug Fixes

• test: import generate_variations directly (ruff N812) (52968ac)

Documentation

• asset tools + commands + provider categories in README/config; roadmap update (c2f3f17)

v0.10.0

0.10.0 (2026-07-07)

Features

• cli: clay texture command + --texture opt-in flag on generate (7d62c41)
• gpu-backend: real /texture route + runtime.generate_texture (honest gating) (8107c1c)
• providers: register texture providers (paint3d/syncmvd/texture) (a577c39)
• texture: TextureAssetGenerator — UV-aware (re)texture + decals + manifest via /texture (81918b8)
• tools: texture_asset tool (re-skin any mesh, GPU-gated) (88cd2bb)

v0.9.0

0.9.0 (2026-07-07)

Features

• cli: clay material command + --material opt-in flag on generate (505e4c2)
• gpu-backend: /material route + runtime.generate_material (honest GPU gating) (0f45970)
• material: MaterialGenerator — PBR set + material.json via /material backend route (b8b9aba)
• providers: provider categories (shape/material/texture) + material providers (31ec59f)
• tools: generate_material tool (5 PBR maps + manifest, GPU-gated) (b8cceba)

v0.8.0

0.8.0 (2026-07-07)

Features

• blender: per-profile auto-rig (biped/quadruped/vehicle/generic) → skinned/parented FBX (5db594a)
• cli: clay rig command + --rig/--rig-type opt-in flags on generate (fa3c9bd)
• tools: rig_asset tool (heuristic per-profile, Blender-backed) (a9eb5f7)

v0.7.0

0.7.0 (2026-07-07)

Features

• blender: high→low tangent-space normal (+AO) bake (script + ops) (085e84b)
• cli: clay bake command + --bake opt-in flag on generate (ce3c63d)
• tools: bake_normals tool (Blender, decimates low if omitted) (01b9b16)

v0.6.0

0.6.0 (2026-07-07)

Features

• blender: shared script helpers + Quadriflow retopo (engine + ops) (8c57b5b)
• cli: clay retopo command + --retopo opt-in flag on generate (d9ba1a9)
• tools: retopo_asset tool (Blender Quadriflow, quad-dominant + re-unwrap) (9a6dcae)

v0.5.0

0.5.0 (2026-07-07)

Features

• cli: clay lods command + --with-lods opt-in flag on generate (54f19eb)
• lods: make_lods — quadric-decimation LOD chain at descending ratios (3c68eb9)
• tools: make_lods tool (69f1332)

v0.4.0

0.4.0 (2026-07-07)

Features

• cli: clay collision command + --collision opt-in flag on generate (476d5f6)
• collision: make_collision — convex/box/simplified/compound (coacd + honest fallback) (8f15584)
• tools: make_collision tool (CPU geometry) (95fc279)

v0.3.0

0.3.0 (2026-07-07)

Features

• blender: FBX export script + ops wrapper (eb29239)
• blender: headless Blender engine — resolve binary, run scripts, honest gating (297934e)
• cli: clay export-fbx command (a3848f9)
• config: [blender] section — headless Blender mesh-processing engine path (39d81ee)
• postprocess: real FBX export via Blender (R1) — replaces honest stub (51fedfe)
• tools: export_fbx tool (Blender-backed, fails visibly) (04da4ae)

Documentation

• config: document [blender] path in example config (c54d99a)

v0.2.0

0.2.0 (2026-07-07)

Features

• bench: default case matrix (tri-budget sweep + object variety) (6755c28)
• bench: Modal harness — real runtime + postprocess, per-case metrics/cost, visible failures (b91daec)
• gpu-backend: multi-arch ext build (8.0;8.6;8.9) — runs on A10G/A100/L40S; default GPU A10G (584860a)
• gpu-backend: real TRELLIS-2 image — from-source CUDA-extension build (nvdiffrast/diffoctreerast/diff-gaussian-rast/vox2seq, xformers/spconv/kaolin) (19b5abb)
• gpu-backend: thread target_tris through the contract to the runtime (9389939)
• preview: clay preview — self-contained interactive model-viewer HTML (2f38238)

Bug Fixes

• bench: use the real shared TRELLIS image (pip install git+ was invalid) (95447d4)
• gpu-backend: --no-build-isolation for CUDA extension builds (nvdiffrast et al.) (6fb66db)
• gpu-backend: add fast-simplification/pygltflib for postprocess (late layer, keeps ext cache) (e122e0f)
• gpu-backend: drive TRELLIS to_glb simplify from tri budget (texture-aware decimation) (b27149a)
• gpu-backend: drop vox2seq (not part of TRELLIS setup); ext build complete (4e5adbd)
• gpu-backend: force gcc/g++ for CUDA ext builds (inline, keeps torch layer cached) (1226f78)
• gpu-backend: install setuptools/wheel before --no-build-isolation ext builds (685c8ca)
• gpu-backend: set ATTN_BACKEND/SPCONV_ALGO before import, run TRELLIS in-place, robust face count (c4742a3)
• postprocess: preserve provider-baked textures — skip destructive re-unwrap/decimate when textured (8e238aa)

Documentation

• bench: first real TRELLIS-2 results on A10G (5/5 textured, ~1.5-11c/asset) (6d43b40)
• bench: how to run the Clay benchmarks (5907e46)
• bench: input image prompts (single-object, Clay-branded, 3D-friendly) (d0c39ad)