blog: llm-d v0.7 release announcement#322
Open
chcost wants to merge 16 commits into
Open
Conversation
Signed-off-by: Carlos H. Andrade Costa <chcost@us.ibm.com>
Add comprehensive v0.7 release blog post covering: - Deployment simplification through standalone mode - Kustomize-native installation across all guides - Hardware expansion (Blackwell, Rebellions, AMD ROCm, TPU v7) - Experimental Flow Control and Batch Gateway features - KV cache management enhancements - P/D disaggregation platform expansion - Documentation and CI overhaul Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com> Signed-off-by: Carlos H. Andrade Costa <chcost@us.ibm.com>
chcost
requested review from
Gregory-Pereira,
clubanderson,
davidgs,
jjasghar,
petecheslock,
robertgshaw2-redhat and
smarterclayton
as code owners
✅ Deploy Preview for elaborate-kangaroo-25e1ee ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
Signed-off-by: Carlos H. Andrade Costa <chcost@us.ibm.com>
Add transformation to strip angle brackets from URLs (e.g., <https://...>) which MDX interprets as HTML tags. This fixes the build failure in artifacts.md pulled from upstream llm-d/llm-d. Signed-off-by: Carlos H. Andrade Costa <chcost@us.ibm.com>
Replace generic forward-looking statements with concrete v0.8 milestone items from issue #1520, organized by theme: - Production graduation (Flow Control, Batch Gateway, multimodal) - RL infrastructure (Python scheduler, non-K8s mode, time-slicing) - Advanced disaggregation (NIXL improvements, new connectors, GB200) - CI/operational rigor (TPU/AMD testing, lm-eval, monitoring) - Deployment flexibility (multi-model, rollout guides) Signed-off-by: Carlos H. Andrade Costa <chcost@us.ibm.com>
Changes for readability and engagement: - Move documentation section to llm-d#2 position (after intro) with prominent stats - Add horizontal rules between major sections for visual breaks - Tighten prose throughout (cut ~40% of words while keeping substance) - Use bold headers and bullet points for scannability - Fix author back to llm-d-maintainers - Compress Looking Ahead to focused bullet points Documentation section now highlights: - 124 files, 10K+ lines as rebuild scale - Architecture reference, visual topology, operational tooling - Interactive scripts and health checks Signed-off-by: Carlos H. Andrade Costa <chcost@us.ibm.com>
- Add llm-d profile logo as avatar image - Remove 'title' field (affiliation) from llm-d-maintainers author Signed-off-by: Carlos H. Andrade Costa <chcost@us.ibm.com>
Major restructuring based on feedback: - Restore longer, more substantive narrative (not bullet-point style) - Consolidate from 10+ fragmented sections into 8 thematic sections - Add performance→operationalization framing in intro - Include external validations (Tesla, Google, AWS, Oracle) with metrics - Remove horizontal rules (too fragmented) - Emphasize shift from proving performance to making it accessible - Use Gemini's section consolidation approach while keeping llm-d voice - Avoid 'lines of code' metrics, focus on contributor/PR volume Signed-off-by: Carlos H. Andrade Costa <chcost@us.ibm.com>
Signed-off-by: Carlos H. Andrade Costa <chcost@us.ibm.com>
Changes: - Use suggested opening: 'If v0.6 was about proving... then v0.7 is about deploying' - Drop commit/PR counts (250 PRs, 157 commits, 23 developers) - Compress performance validation paragraph to one line with key examples - Focus on impact rather than process metrics Signed-off-by: Carlos H. Andrade Costa <chcost@us.ibm.com>
Replace three separate performance validation links with one link to the performance-highlights section on GitHub README Signed-off-by: Carlos H. Andrade Costa <chcost@us.ibm.com>
Changes: - Remove '124 changed markdown files', '72 new files', '164 files' - Keep only 'over 10,000 lines of new content' as scale indicator - Change 'Fifteen new SVG diagrams' to simply 'New diagrams' - Focus on impact rather than counting artifacts Signed-off-by: Carlos H. Andrade Costa <chcost@us.ibm.com>
Let the performance highlights link speak for itself without listing specific numbers (3×, 70%, 40%) Signed-off-by: Carlos H. Andrade Costa <chcost@us.ibm.com>
Add opening line highlighting documentation as largest category by volume Signed-off-by: Carlos H. Andrade Costa <chcost@us.ibm.com>
Signed-off-by: Carlos H. Andrade Costa <chcost@us.ibm.com>
…tainers' Signed-off-by: Carlos H. Andrade Costa <chcost@us.ibm.com>
Contributor
Author
|
@robertgshaw2-redhat @ahg-g good with this one? |
Collaborator
|
I think we should mention the llm-d Router terminology change, see llm-d/llm-d#1298 for content. |
Collaborator
|
otherwise, looks good to me |
Contributor
Author
|
ah, good catch. Let me add. Thx |
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
3 participants
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.
Add comprehensive v0.7 release blog post covering the operational maturity focus of this release.
Preview the blog post formatting once the PR is created through the Netlify deploy preview.