Skip to content

Pull requests: hao-ai-lab/FastVideo

Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort

Pull requests list

[docs]: document automatic Docker image builds scope: docs Documentation type: docs Documentation only
#1608 opened Jul 15, 2026 by SolitaryThinker Collaborator Loading…
[bugfix] reason1: handle BatchEncoding from apply_chat_template scope: infra CI, tests, Docker, build scope: model Model architecture (DiTs, encoders, VAEs) type: bugfix Bug fix
#1607 opened Jul 15, 2026 by Mister-Raggs Contributor Loading…
[bugfix] Use pytest pythonpath config so scoped invocations can find fastvideo scope: infra CI, tests, Docker, build type: bugfix Bug fix
#1605 opened Jul 15, 2026 by yogyam Loading…
2 tasks done
[perf]: enable per-block torch.compile for LTX2 with persistent cache scope: attention Attention backends (VSA, STA, Flash, etc.) scope: model Model architecture (DiTs, encoders, VAEs) type: perf Performance improvement
#1602 opened Jul 14, 2026 by Someone45 Loading…
[feat] Add Kandinsky5 QAD training pipeline: data preprocessing, QAT finetune, QAT-aware DMD distillation scope: data Data preprocessing, datasets scope: inference Inference pipeline, serving, CLI scope: infra CI, tests, Docker, build scope: model Model architecture (DiTs, encoders, VAEs) scope: training Training pipeline, methods, configs type: feat New feature or capability
#1601 opened Jul 14, 2026 by leffff Contributor Loading…
[new-model] Add LingBot-Video TI2V inference scope: inference Inference pipeline, serving, CLI scope: infra CI, tests, Docker, build scope: model Model architecture (DiTs, encoders, VAEs) type: new-model New model support
#1600 opened Jul 14, 2026 by Davids048 Collaborator Draft
[kernel] Build + allow attn_qat_infer FP4 attention on sm_121a (DGX Spark) ready PR is ready to merge scope: attention Attention backends (VSA, STA, Flash, etc.) scope: docs Documentation scope: inference Inference pipeline, serving, CLI scope: infra CI, tests, Docker, build scope: kernel CUDA kernels, fastvideo-kernel scope: model Model architecture (DiTs, encoders, VAEs)
#1598 opened Jul 13, 2026 by Mister-Raggs Contributor Loading…
[new-model] Add LingBot-Video Dense and MoE/refiner T2V inference needs-rebase PR has merge conflicts scope: docs Documentation scope: inference Inference pipeline, serving, CLI scope: infra CI, tests, Docker, build scope: model Model architecture (DiTs, encoders, VAEs) type: new-model New model support
#1595 opened Jul 13, 2026 by Davids048 Collaborator Loading…
[perf] LTX2 FP4-linear on DGX Spark (GB10/sm_121): ~24% faster denoise at 1080p scope: inference Inference pipeline, serving, CLI type: perf Performance improvement
#1594 opened Jul 13, 2026 by Mister-Raggs Contributor Loading…
[ci]: pin VSA training regression to H100 scope: infra CI, tests, Docker, build type: ci CI/CD infrastructure
#1591 opened Jul 12, 2026 by macthecadillac Collaborator Loading…
4 of 6 tasks
[ci]: harden Modal repository checkout ready PR is ready to merge scope: infra CI, tests, Docker, build type: ci CI/CD infrastructure
#1590 opened Jul 12, 2026 by macthecadillac Collaborator Loading…
4 tasks done
[ci]: add transient pytest reruns for Modal tests scope: docs Documentation scope: infra CI, tests, Docker, build type: ci CI/CD infrastructure
#1581 opened Jul 10, 2026 by macthecadillac Collaborator Loading…
4 tasks done
[new-model] Port LingBot-World-v2 scope: docs Documentation scope: inference Inference pipeline, serving, CLI scope: infra CI, tests, Docker, build scope: model Model architecture (DiTs, encoders, VAEs) type: new-model New model support
#1579 opened Jul 10, 2026 by Davids048 Collaborator Loading…
[ci]: cover transformers 5.x config compat for Qwen2.5-VL (#1576) scope: infra CI, tests, Docker, build type: ci CI/CD infrastructure
#1577 opened Jul 10, 2026 by Mister-Raggs Contributor Loading…
[bugfix] Skip CUDA-only fastvideo-kernel/flashinfer-python deps on non-Linux scope: infra CI, tests, Docker, build type: bugfix Bug fix
#1574 opened Jul 9, 2026 by yogyam Loading…
4 of 5 tasks
[bugfix]: skip unused output materialization ready PR is ready to merge scope: inference Inference pipeline, serving, CLI scope: infra CI, tests, Docker, build type: bugfix Bug fix
#1567 opened Jul 7, 2026 by macthecadillac Collaborator Loading…
4 of 6 tasks
[bugfix]: guard FastWan FullAttn backend config scope: attention Attention backends (VSA, STA, Flash, etc.) scope: docs Documentation scope: inference Inference pipeline, serving, CLI scope: infra CI, tests, Docker, build scope: model Model architecture (DiTs, encoders, VAEs) scope: training Training pipeline, methods, configs type: bugfix Bug fix
#1563 opened Jul 6, 2026 by macthecadillac Collaborator Loading…
5 of 6 tasks
[ci]: cache FastVideo kernel builds in Modal scope: infra CI, tests, Docker, build scope: kernel CUDA kernels, fastvideo-kernel type: ci CI/CD infrastructure
#1562 opened Jul 6, 2026 by macthecadillac Collaborator Loading…
4 of 6 tasks
[fix] Fix latent handoff and FastWan-QAD example path needs-rebase PR has merge conflicts scope: inference Inference pipeline, serving, CLI scope: infra CI, tests, Docker, build type: bugfix Bug fix
#1557 opened Jul 5, 2026 by aryan5v Contributor Loading…
[feat] train: add Muon optimizer (FSDP2/DTensor-aware) alongside AdamW scope: infra CI, tests, Docker, build scope: training Training pipeline, methods, configs type: feat New feature or capability
#1519 opened Jun 30, 2026 by alexzms Collaborator Draft
2 of 3 tasks
[feat] Add InterleaveThinker inference workflow scope: docs Documentation scope: inference Inference pipeline, serving, CLI scope: infra CI, tests, Docker, build type: feat New feature or capability
#1509 opened Jun 29, 2026 by macthecadillac Collaborator Loading…
6 tasks done
[misc] Clean up QAD 5090 example inference scripts scope: inference Inference pipeline, serving, CLI scope: kernel CUDA kernels, fastvideo-kernel type: misc Cleanup, config, dependencies
#1496 opened Jun 25, 2026 by kevin314 Collaborator Loading…
1 of 6 tasks
[feat]: hard-fail on missing/mismatched attention backends; require VSA for FastWan ready PR is ready to merge scope: attention Attention backends (VSA, STA, Flash, etc.) scope: docs Documentation scope: inference Inference pipeline, serving, CLI scope: infra CI, tests, Docker, build scope: model Model architecture (DiTs, encoders, VAEs) scope: training Training pipeline, methods, configs type: feat New feature or capability
#1494 opened Jun 25, 2026 by SolitaryThinker Collaborator Loading…
[feat] Add DiffusionNFT Wan video RL training scope: data Data preprocessing, datasets scope: infra CI, tests, Docker, build scope: training Training pipeline, methods, configs type: feat New feature or capability
#1476 opened Jun 22, 2026 by Abecid Loading…
3 of 6 tasks
[docs] Add GenRL asset preparation recipe scope: training Training pipeline, methods, configs type: docs Documentation only
#1456 opened Jun 12, 2026 by Abecid Loading…
3 of 4 tasks
ProTip! Adding no:label will show everything without a label.