Repository navigation
docs(use-cases): use case guides refactor - #16607
Merged
Merged
Conversation
Restructure the Use Cases tab so every section is one level deep: Overview, Reinforcement Learning, Agents, Image, Video, Speech & Audio, Multimodal Performance, and Many-Model Serving. - Split Multimodal Serving and Diffusion into Image, Video, and Speech & Audio sections that cover both understanding and generation, plus a Multimodal Performance section for the shared image/video features. - Add Image Understanding, Video Understanding, Speech Recognition, Audio Understanding, Speech-to-Speech, a Use Cases overview, a Multimodal Performance overview, a Many-Model overview, and a Common Setup page. - Keep backend setup in Use Cases: "Set up the backend" tabs on the first page of each section, shared through _snippets/. - Split FastVideo, MiniMax-H3, and exact video KV routing into their own pages; merge Additional Media Decoders into Video Decoding. - Reorder RL (integrations first) and Agents (harness, session, hints, tracing, replay, ThunderAgent, priority). - Move Tool Calling & Reasoning to Developer Guide > Frontend and LLM diffusion to Developer Guide > Backends > SGLang. - Add dev-scoped redirects for every moved URL and retarget existing ones. P0 fixes: agent tool-event wire format, replay segment file name, agent harness trace path, TensorRT-LLM text-to-video model name, TTS streaming and voice cloning, SGLang image-to-video request, TensorRT-LLM E/PD support, missing tool-call parsers, stale parser crate version. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: harryk <harryk@nvidia.com> Signed-off-by: Harry Kim <harryk@nvidia.com> Signed-off-by: akshatha-k <akshutk@gmail.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Harry Kim <harryk@nvidia.com> Signed-off-by: akshatha-k <akshutk@gmail.com>
Put the MiniMax-H3 (T2VA) tutorial ahead of the standard text-to-video guide on one page, with a tutorials table at the top linking to both. Rename the pages Text to Video (T2V) and Image to Video (I2V), keeping their URLs. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Harry Kim <harryk@nvidia.com> Signed-off-by: akshatha-k <akshutk@gmail.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Harry Kim <harryk@nvidia.com> Signed-off-by: akshatha-k <akshutk@gmail.com>
…page Upstream ai-dynamo#16350 added a link to the old diffusion/workflows/text-to-image.md path, which the Use Cases restructure moves to image/image-generation.md. Signed-off-by: akshatha-k <akshutk@gmail.com>
…ith Speech to Text Merge the two speech pages into one Speech to Text guide and verify its content against the vLLM realtime transcription handler, the frontend /v1/realtime route, the launch script and client, and the Nemotron ASR adapter. The launcher defaults --hf-overrides for the default model, the worker accepts 24 kHz PCM16 and English only, and the guide documents the session limits, error events, and the Nemotron adapter's differences. Drop the Audio Understanding page and its unverified support matrix, and point readers at the vLLM multimodal guide for audio_url requests. Signed-off-by: akshatha-k <akshutk@gmail.com>
akshatha-k
temporarily deployed
to
external_collaborator
October 16, 2026 00:17 — with
GitHub Actions
Inactive
akshatha-k
temporarily deployed
to
external_collaborator
October 16, 2026 00:17 — with
GitHub Actions
Inactive
Contributor
Contributor
|
/ok to test d494bfb |
…ections Restructure the Use Cases tab: - Rename the Image and Video sections to Image Understanding and Video Understanding. Each starts with an Overview page, and ends with a Performance Optimizations section that links to the VLM page. - Remove the Multimodal Performance section. Add "VLM Performance Optimization" under Image Understanding with an end-to-end VLM request flow diagram, a table of when each feature helps, and collapsible sections for multimodal KV routing, frontend media decoding, the shared image cache, the embedding cache, and encoder disaggregation. - Keep Custom Vision Encoders as a page inside Image Understanding. - Move generation content (image generation, text to video, image to video, FastVideo) into a new top-level Diffusion section with an overview page. - Update nav, links, and redirects for every moved page. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> Signed-off-by: akshatha-k <akshutk@gmail.com>
akshatha-k
temporarily deployed
to
external_collaborator
October 16, 2026 00:57 — with
GitHub Actions
Inactive
Contributor
|
/ok to test 5cf311a |
Contributor
✅ Dynamo PR CI passed — run 42105782614 (attempt 1) on
|
Use the NVIDIA green and white palette with blue highlights on a white background, and move the image to docs/fern/assets/img/ per the documentation style guide. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> Signed-off-by: akshatha-k <akshutk@gmail.com>
akshatha-k
temporarily deployed
to
external_collaborator
October 16, 2026 17:31 — with
GitHub Actions
Inactive
Contributor
|
/ok to test d49a460 |
… diagram - Rename Image Generation to Text to Image (file, URL, nav, links, and redirects) to match Text to Video. - Put FastVideo right below the Diffusion overview and add a pointer to it at the top of Text to Video. - Drop the Diffusion overview intro so it opens with the guides table. - Rename the "Speech and Audio" label to "Speech". - Reword the VLM Performance Optimization intro. - Redraw the VLM request path diagram with the pipeline in blue and the optional features in green with dotted arrows. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> Signed-off-by: akshatha-k <akshutk@gmail.com>
akshatha-k
temporarily deployed
to
external_collaborator
October 16, 2026 17:46 — with
GitHub Actions
Inactive
Contributor
|
/ok to test 363e458 |
…rder sections - Revert kv-cache-offloading and agent-harnesses/simulation/tracing pages to main - Move Additional Media Decoders to Developer Guide > Developing and Extending Dynamo - Collapse GPU Sharing into one overview page with setup, experimental benchmark results, and a how-to-read accordion; move the deploy and MPS guides to examples/gpu-sharing/docs and link them from the README - Add redirects for the old many-model-serving pages - Move Speech above Diffusion in the Use Cases nav and overview table - Remove the Deployment Notation section from VLM Performance Optimization Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> Signed-off-by: akshatha-k <akshutk@gmail.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> Signed-off-by: akshatha-k <akshutk@gmail.com>
Contributor
|
/ok to test ac7ce05 |
Use one green accent for components, a single arrow color, gray for paths not taken, and amber only for state. Replace generic icons with component-specific ones and drop the diagram eyebrow labels. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> Signed-off-by: akshatha-k <akshutk@gmail.com>
akshatha-k
temporarily deployed
to
external_collaborator
November 2, 2026 07:09 — with
GitHub Actions
Inactive
Contributor
|
/ok to test 1c29d8f |
Resolve conflicts with main and carry main's doc changes into the reorganized Use Cases layout. - Keep main's prose in developer-guide and reference pages and point their links at the moved use-case pages. - Port main's edits into pages this branch moved: EDF default queue policy and the flat policy-class schema (priority scheduling), SGLang wide-token routing fallback (image and video KV routing), and Cosmos3 / rejected file:// references. - Restore main's Diffusion overview and workflow pages, with Workflows open by default and FastVideo as the first Text-to-Video tab. Text-to-Speech stays under Speech. - Convert GitHub-style callouts on the Diffusion pages to Fern components so they render in previews. - Fix links main added to moved paths, and realign redirects so every diffusion redirect lands on a live page. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: akshatha-k <akshutk@gmail.com>
akshatha-k
temporarily deployed
to
external_collaborator
November 5, 2026 19:20 — with
GitHub Actions
Inactive
Contributor
|
/ok to test 954aa1a |
Bring the branch up to date with main (planner retune scaling policy, batch gateway docs). No conflicts. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: akshatha-k <akshutk@gmail.com>
akshatha-k
temporarily deployed
to
external_collaborator
November 5, 2026 19:22 — with
GitHub Actions
Inactive
Contributor
|
/ok to test bfc4463 |
…rser entries Main already lists kimi_k3 and deepseek_v41 with fuller notes; keep those and fold in the recipe links from the duplicates. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: akshatha-k <akshutk@gmail.com>
akshatha-k
temporarily deployed
to
external_collaborator
November 5, 2026 20:41 — with
GitHub Actions
Inactive
Contributor
|
/ok to test 7f36999 |
Jont828
added a commit
to Jont828/dynamo
that referenced
this pull request
Nov 6, 2026
Rebasing onto main picked up the use case guides refactor (ai-dynamo#16607), which moved or replaced pages this PR links to or renames into: - multimodal-serving/voice.md was replaced by speech/voice-pipelines.md. The Speech Recognition on Kubernetes section now lives there (ported during conflict resolution), so point the vLLM multimodal guide and examples/backends/vllm/deploy/README.md at the new page and anchors. - multimodal-serving/additional-media-decoders.md moved to image-understanding/additional-media-decoders.md; update the link in the vLLM multimodal guide. - The new speech/speech-to-text.md page launches the realtime ASR worker with agg_realtime_transcription.sh, which this PR renames to agg_asr_realtime.sh; apply the same rename there. Automated-by: pr-shepherd Signed-off-by: Jonathan Tong <jt572@cornell.edu>
Jont828
added a commit
to Jont828/dynamo
that referenced
this pull request
Nov 6, 2026
Rebasing onto main picked up the use case guides refactor (ai-dynamo#16607), which moved pages that the migrated example pages link to: - diffusion/workflows/text-to-audio.md moved to speech/text-to-speech.md. Main updated the Audex link in recipes/cli-templates/vllm.mdx, which this PR removes; port that change to audio-generation.mdx, the page that now carries the Qwen3-TTS and Audex guidance, and use the new page title as link text. - use-cases/gpu-sharing/gpu-sharing-results.mdx was folded into use-cases/gpu-sharing/overview.mdx, and the deploy-*.md guides moved out of the docs site to examples/gpu-sharing/docs/. Point gpu-sharing.mdx at the overview page and at the guides on GitHub, as the overview page does. Automated-by: pr-shepherd Signed-off-by: Jonathan Tong <jt572@cornell.edu>
ishandhanani
added a commit
to ishandhanani/dynamo
that referenced
this pull request
Nov 6, 2026
Twenty-six upstream commits. Conflicts in eight paths, resolved into the owner crates; everything else merged by path. - ai-dynamo#17910 strict boolean worker flags: lib/router/scheduler/src/selection/types.rs (kernel WorkerRequest and WorkerPatchRequest take strict_bool on is_eagle and engine_owned_mm_identity; carried by rename detection), lib/bindings/python/tests/test_kv_bindings.py. - ai-dynamo#18240 configurable session-affinity entry limit: lib/llm/src/session_affinity/mod.rs reads SessionAffinityConfig and MAX_SESSION_AFFINITY_ENTRIES from dynamo_session_affinity::host (the host contract now exports the entry limit), coordinator.rs and kv_router/embedded.rs build the table through affinity_config; frontend args, env names, docs and Python tests auto-merged. - ai-dynamo#18145 EPP co-ownership: .github/codeowners/areas.yaml globs re-pointed to the moved files (lib/router/kv-router/src/plugins/registry.rs; lib/router/kv-router/src/scheduling/config.rs plus the kernel slice lib/router/scheduler/src/config.rs; the pre-split services/selection top level as lib/router/scheduler/src/selection/*, lib/router/kv-router/src/services/selection/* and lib/router/session-affinity/src/lib.rs, core/ stays router-only); the lib/llm, router-policies and example globs auto-merged; CODEOWNERS regenerated. - ai-dynamo#17708 nvtx in pre-merge CI: .github/workflows/pre-merge.yml gains "Run NVTX feature tests" after the stack's kv-router adapter and canonical replay steps and "Lint the NVTX feature" in rust-clippy; Cargo.toml profiling comment, pyproject.toml. - ai-dynamo#17623 dependency bumps: lib/mocker/Cargo.toml adds zip next to the stack's fastrand; lib/llm/Cargo.toml utoipa-swagger-ui 10.0.2; the root, python-bindings and kvbm lockfiles taken from upstream and regenerated. - ai-dynamo#17675 DynamoDB-backed global session affinity: lib/global-router, lib/global-frontend, new lib/global-router-store-dynamodb workspace member. - ai-dynamo#18278 synthetic session prompts in replay: lib/mocker/src/replay/entrypoints.rs, lib/bindings/python/rust/llm/replay.rs, _core.pyi, components replay package and tests. - ai-dynamo#16607 use-case docs refactor: docs/fern; multimodal-kv-routing.md moved to use-cases/image-understanding/vlm-performance-optimization/ and the stack's DYN_LOG directive (dynamo_scheduler, dynamo_admission, dynamo_worker_selection) carried to the new path. - ai-dynamo#17706 NVTX ranges on the OpenAI HTTP path, ai-dynamo#17626 beam tool constraints, ai-dynamo#18015 MiniMax SSE keep-alive: lib/llm, auto-merged. - ai-dynamo#18229 Velo in-place encoding: lib/runtime. ai-dynamo#17370 SGLang first-chunk logprobs: lib/generate-formats. - ai-dynamo#17862 vLLM 0.33.0 bump: container, components, lib/sidecar/vllm, pyproject. - ai-dynamo#18121, ai-dynamo#18154, ai-dynamo#18111 planner live retune; ai-dynamo#18234 Hunyuan reasoning defaults; ai-dynamo#17229 TRT-LLM KVBM removal: components, tests, bindings Python, docs. - ai-dynamo#18069, ai-dynamo#18066 operator; ai-dynamo#18244 recipes; ai-dynamo#18210 batch docs; three automated community-events refreshes. Signed-off-by: Ishan Dhanani <ishandhanani@gmail.com>
Jont828
added a commit
to Jont828/dynamo
that referenced
this pull request
Nov 6, 2026
Rebasing onto main picked up the use case guides refactor (ai-dynamo#16607), which moved pages that the migrated example pages link to: - diffusion/workflows/text-to-audio.md moved to speech/text-to-speech.md. Main updated the Audex link in recipes/cli-templates/vllm.mdx, which this PR removes; port that change to audio-generation.mdx, the page that now carries the Qwen3-TTS and Audex guidance, and use the new page title as link text. - use-cases/gpu-sharing/gpu-sharing-results.mdx was folded into use-cases/gpu-sharing/overview.mdx, and the deploy-*.md guides moved out of the docs site to examples/gpu-sharing/docs/. Point gpu-sharing.mdx at the overview page and at the guides on GitHub, as the overview page does. Automated-by: pr-shepherd Signed-off-by: Jonathan Tong <jt572@cornell.edu>
Jont828
added a commit
to Jont828/dynamo
that referenced
this pull request
Nov 6, 2026
…hanges Adjustments for upstream changes that landed after this branch's old merge base and that the rebase does not catch as conflicts: - Qwen3-ASR example (ai-dynamo#15671) added deploy/agg_asr.yaml, deploy/agg_asr_realtime.yaml, and launch/agg_asr.sh under examples/backends/vllm, and renamed launch/agg_realtime_transcription.sh to agg_asr_realtime.sh. Embed the three new sources on the Audio and Realtime Serving page, with the audio-transfer and image notes upstream added to the vLLM deploy README, follow the rename (also the change upstream made in the removed vLLM CLI template), and update the card keywords on the overview and Workloads pages. - The use-case refactor (ai-dynamo#16607) moved diffusion/workflows/text-to-audio.md to speech/text-to-speech.md. Point the Audio Generation links, and their text, at the new page. - ai-dynamo#16607 moved the GPU sharing walkthroughs from use-cases/gpu-sharing/ into examples/gpu-sharing/docs/ and replaced gpu-sharing-results.mdx with overview.mdx. Retarget the GPU Sharing example page, linking the walkthroughs on GitHub, and describe the README as this branch now shapes it: a file list that links each configuration's guide. - ai-dynamo#16607 deleted use-cases/multimodal-serving/voice.md and moved its content to speech/voice-pipelines.md and speech/speech-to-text.md. Carry over this branch's change to voice.md: point the Nemotron deployment and external-orchestration links at the migrated Nemotron Speech Pipeline page instead of README anchors this branch removes. - ai-dynamo#18210 added a concurrency and result-delivery tuning section to the batch-gateway README, whose content this branch moved to offline-batch-inference.md. Port the section there, with Title Case headings to match the page and the working directory stated as elsewhere on the page. Signed-off-by: Jonathan Tong <jt572@cornell.edu> Automated-by: pr-shepherd
Jont828
added a commit
to Jont828/dynamo
that referenced
this pull request
Nov 9, 2026
…hanges Adjustments for upstream changes that landed after this branch's old merge base and that the rebase does not catch as conflicts: - Qwen3-ASR example (ai-dynamo#15671) added deploy/agg_asr.yaml, deploy/agg_asr_realtime.yaml, and launch/agg_asr.sh under examples/backends/vllm, and renamed launch/agg_realtime_transcription.sh to agg_asr_realtime.sh. Embed the three new sources on the Audio and Realtime Serving page, with the audio-transfer and image notes upstream added to the vLLM deploy README, follow the rename (also the change upstream made in the removed vLLM CLI template), and update the card keywords on the overview and Workloads pages. - The use-case refactor (ai-dynamo#16607) moved diffusion/workflows/text-to-audio.md to speech/text-to-speech.md. Point the Audio Generation links, and their text, at the new page. - ai-dynamo#16607 moved the GPU sharing walkthroughs from use-cases/gpu-sharing/ into examples/gpu-sharing/docs/ and replaced gpu-sharing-results.mdx with overview.mdx. Retarget the GPU Sharing example page, linking the walkthroughs on GitHub, and describe the README as this branch now shapes it: a file list that links each configuration's guide. - ai-dynamo#16607 deleted use-cases/multimodal-serving/voice.md and moved its content to speech/voice-pipelines.md and speech/speech-to-text.md. Carry over this branch's change to voice.md: point the Nemotron deployment and external-orchestration links at the migrated Nemotron Speech Pipeline page instead of README anchors this branch removes. - ai-dynamo#18210 added a concurrency and result-delivery tuning section to the batch-gateway README, whose content this branch moved to offline-batch-inference.md. Port the section there, with Title Case headings to match the page and the working directory stated as elsewhere on the page. Signed-off-by: Jonathan Tong <jt572@cornell.edu> Automated-by: pr-shepherd
Jont828
added a commit
to Jont828/dynamo
that referenced
this pull request
Nov 10, 2026
…hanges Adjustments for upstream changes that landed after this branch's old merge base and that the rebase does not catch as conflicts: - Qwen3-ASR example (ai-dynamo#15671) added deploy/agg_asr.yaml, deploy/agg_asr_realtime.yaml, and launch/agg_asr.sh under examples/backends/vllm, and renamed launch/agg_realtime_transcription.sh to agg_asr_realtime.sh. Embed the three new sources on the Audio and Realtime Serving page, with the audio-transfer and image notes upstream added to the vLLM deploy README, follow the rename (also the change upstream made in the removed vLLM CLI template), and update the card keywords on the overview and Workloads pages. - The use-case refactor (ai-dynamo#16607) moved diffusion/workflows/text-to-audio.md to speech/text-to-speech.md. Point the Audio Generation links, and their text, at the new page. - ai-dynamo#16607 moved the GPU sharing walkthroughs from use-cases/gpu-sharing/ into examples/gpu-sharing/docs/ and replaced gpu-sharing-results.mdx with overview.mdx. Retarget the GPU Sharing example page, linking the walkthroughs on GitHub, and describe the README as this branch now shapes it: a file list that links each configuration's guide. - ai-dynamo#16607 deleted use-cases/multimodal-serving/voice.md and moved its content to speech/voice-pipelines.md and speech/speech-to-text.md. Carry over this branch's change to voice.md: point the Nemotron deployment and external-orchestration links at the migrated Nemotron Speech Pipeline page instead of README anchors this branch removes. - ai-dynamo#18210 added a concurrency and result-delivery tuning section to the batch-gateway README, whose content this branch moved to offline-batch-inference.md. Port the section there, with Title Case headings to match the page and the working directory stated as elsewhere on the page. Signed-off-by: Jonathan Tong <jt572@cornell.edu> Automated-by: pr-shepherd
Jont828
added a commit
to Jont828/dynamo
that referenced
this pull request
Nov 11, 2026
…hanges Adjustments for upstream changes that landed after this branch's old merge base and that the rebase does not catch as conflicts: - Qwen3-ASR example (ai-dynamo#15671) added deploy/agg_asr.yaml, deploy/agg_asr_realtime.yaml, and launch/agg_asr.sh under examples/backends/vllm, and renamed launch/agg_realtime_transcription.sh to agg_asr_realtime.sh. Embed the three new sources on the Audio and Realtime Serving page, with the audio-transfer and image notes upstream added to the vLLM deploy README, follow the rename (also the change upstream made in the removed vLLM CLI template), and update the card keywords on the overview and Workloads pages. - The use-case refactor (ai-dynamo#16607) moved diffusion/workflows/text-to-audio.md to speech/text-to-speech.md. Point the Audio Generation links, and their text, at the new page. - ai-dynamo#16607 moved the GPU sharing walkthroughs from use-cases/gpu-sharing/ into examples/gpu-sharing/docs/ and replaced gpu-sharing-results.mdx with overview.mdx. Retarget the GPU Sharing example page, linking the walkthroughs on GitHub, and describe the README as this branch now shapes it: a file list that links each configuration's guide. - ai-dynamo#16607 deleted use-cases/multimodal-serving/voice.md and moved its content to speech/voice-pipelines.md and speech/speech-to-text.md. Carry over this branch's change to voice.md: point the Nemotron deployment and external-orchestration links at the migrated Nemotron Speech Pipeline page instead of README anchors this branch removes. - ai-dynamo#18210 added a concurrency and result-delivery tuning section to the batch-gateway README, whose content this branch moved to offline-batch-inference.md. Port the section there, with Title Case headings to match the page and the working directory stated as elsewhere on the page. Signed-off-by: Jonathan Tong <jt572@cornell.edu> Automated-by: pr-shepherd
Jont828
added a commit
to Jont828/dynamo
that referenced
this pull request
Nov 15, 2026
…hanges Adjustments for upstream changes that landed after this branch's old merge base and that the rebase does not catch as conflicts: - Qwen3-ASR example (ai-dynamo#15671) added deploy/agg_asr.yaml, deploy/agg_asr_realtime.yaml, and launch/agg_asr.sh under examples/backends/vllm, and renamed launch/agg_realtime_transcription.sh to agg_asr_realtime.sh. Embed the three new sources on the Audio and Realtime Serving page, with the audio-transfer and image notes upstream added to the vLLM deploy README, follow the rename (also the change upstream made in the removed vLLM CLI template), and update the card keywords on the overview and Workloads pages. - The use-case refactor (ai-dynamo#16607) moved diffusion/workflows/text-to-audio.md to speech/text-to-speech.md. Point the Audio Generation links, and their text, at the new page. - ai-dynamo#16607 moved the GPU sharing walkthroughs from use-cases/gpu-sharing/ into examples/gpu-sharing/docs/ and replaced gpu-sharing-results.mdx with overview.mdx. Retarget the GPU Sharing example page, linking the walkthroughs on GitHub, and describe the README as this branch now shapes it: a file list that links each configuration's guide. - ai-dynamo#16607 deleted use-cases/multimodal-serving/voice.md and moved its content to speech/voice-pipelines.md and speech/speech-to-text.md. Carry over this branch's change to voice.md: point the Nemotron deployment and external-orchestration links at the migrated Nemotron Speech Pipeline page instead of README anchors this branch removes. - ai-dynamo#18210 added a concurrency and result-delivery tuning section to the batch-gateway README, whose content this branch moved to offline-batch-inference.md. Port the section there, with Title Case headings to match the page and the working directory stated as elsewhere on the page. Signed-off-by: Jonathan Tong <jt572@cornell.edu> Automated-by: pr-shepherd
Jont828
added a commit
to Jont828/dynamo
that referenced
this pull request
Nov 17, 2026
…hanges Adjustments for upstream changes that landed after this branch's old merge base and that the rebase does not catch as conflicts: - Qwen3-ASR example (ai-dynamo#15671) added deploy/agg_asr.yaml, deploy/agg_asr_realtime.yaml, and launch/agg_asr.sh under examples/backends/vllm, and renamed launch/agg_realtime_transcription.sh to agg_asr_realtime.sh. Embed the three new sources on the Audio and Realtime Serving page, with the audio-transfer and image notes upstream added to the vLLM deploy README, follow the rename (also the change upstream made in the removed vLLM CLI template), and update the card keywords on the overview and Workloads pages. - The use-case refactor (ai-dynamo#16607) moved diffusion/workflows/text-to-audio.md to speech/text-to-speech.md. Point the Audio Generation links, and their text, at the new page. - ai-dynamo#16607 moved the GPU sharing walkthroughs from use-cases/gpu-sharing/ into examples/gpu-sharing/docs/ and replaced gpu-sharing-results.mdx with overview.mdx. Retarget the GPU Sharing example page, linking the walkthroughs on GitHub, and describe the README as this branch now shapes it: a file list that links each configuration's guide. - ai-dynamo#16607 deleted use-cases/multimodal-serving/voice.md and moved its content to speech/voice-pipelines.md and speech/speech-to-text.md. Carry over this branch's change to voice.md: point the Nemotron deployment and external-orchestration links at the migrated Nemotron Speech Pipeline page instead of README anchors this branch removes. - ai-dynamo#18210 added a concurrency and result-delivery tuning section to the batch-gateway README, whose content this branch moved to offline-batch-inference.md. Port the section there, with Title Case headings to match the page and the working directory stated as elsewhere on the page. Signed-off-by: Jonathan Tong <jt572@cornell.edu> Automated-by: pr-shepherd
This branch was previously deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Closes DGH-2446
Overview
Restructures the Use Cases tab by workload and modality (Image Understanding, Video Understanding, Speech, Diffusion, GPU Sharing) and rewrites the Agents guides as task-based guides. Opened as a draft to get the Fern preview.
Details
1. Navigation and structure
/diffusion/text-to-image).2. New and consolidated pages
examples/gpu-sharing/docs/and are linked from that README.3. Agents guides rewritten as task-based guides
--router-queue-thresholdand--router-queue-policyflags.session_id/parent_session_id(nottrajectory_id),DYN_REQUEST_TRACE_FILE_PATHreplaces the deprecatedDYN_REQUEST_TRACE_OUTPUT_PATH, and trace segments are namedrequest-trace.000000.jsonl.gz.session-i-ds,agent-hints, andpriority-schedulingURLs.4. Content fixes
--served-model-name wan_t2v.input_referencerequest.kimi_k3,deepseek_v41, andmuse_glimmer. A staledynamo-parsers-v2paragraph is removed.5. Moves out of Use Cases
6. Links, redirects, assets
docs: redirect dead docs paths linked from published versions). One entry, the v1.2.0priority-schedulingredirect, is retargeted to the renamed Prioritize Requests page./many-model-serving/*redirects to/gpu-sharing. Redirect sources that are now real pages are removed.lib/llm/src/preprocessor/media/README.md.docs/fern/assets/img/vlm-request-path.svg.Where the reviewer should start
docs/fern/index.yml(Use Cases tab)docs/fern/pages/use-cases/image-understanding/vlm-performance-optimization.mdand the five pages undervlm-performance-optimization/docs/fern/pages/use-cases/gpu-sharing/overview.mdxandexamples/gpu-sharing/docs/fern/pages/use-cases/agents/overview.mdandrun-an-agent-on-dynamo.mddocs/fern/docs.yml(redirect blocks)Validation
python3 docs/fern/scripts/docs_lint.py: 0 errors.fern check: 0 errors.fern docs broken-links: 1 error inpages/recipes/model-recipes/deepseek-v4-pro-0813.mdx, which this PR does not touch.Needs owner review
<image-to-video-model>placeholder; the repo names no supported checkpoint.Qwen/Qwen3-TTS-12Hz-1.7B-Base(name taken from a test fixture).🤖 Generated with Claude Code