docs(ENG-EXPERT-STREAM): scope the harness disclosure to the runs it is true of - #1444
Merged
Conversation
…is true of The binary disclosure said every figure below came from the W0e C-ABI harness. The streaming-off row does not: the 16 August 2026 run served through vllm-server, which is the binary the recipe publishes. Correcting one undisclosed difference by overstating it is the same defect in the other direction, so the sentence now names the W0e and W0f figures and names the exception. FOLLOWING_AGENTS_PROTOCOL Following-Agents-Protocol: true AI-Assisted: true Assisted-by: AGENT:claude-opus-5 [Claude Code]
The harness-scope correction was authored against #1211 and did not land with it, so it arrived without an issue. #1447 now tracks it, and this appends its index row and cites it from the spec paragraph the correction rewrites, so the three surfaces that must agree do. FOLLOWING_AGENTS_PROTOCOL Following-Agents-Protocol: true AI-Assisted: true Assisted-by: AGENT:claude-opus-5 [Claude Code]
`check-agent-record.py` refused the row at `.agents/issue-index.md:478` with "table has 4 pipes; expected 5". The table is `| Issue | Row | Title | Kind |` and the row I appended stopped after the title. Kind is `record`, which is what the index uses for a defect in a record rather than in code. The checker caught this on the commit that introduced it, and the repair is the row rather than the baseline. FOLLOWING_AGENTS_PROTOCOL Following-Agents-Protocol: true AI-Assisted: true Assisted-by: AGENT:claude-opus-5 [Claude Code]
localai-bot
pushed a commit
that referenced
this pull request
Aug 20, 2026
`origin/main` moved to `f07f96e1c` while this pull request was open, and GitHub reported the branch CONFLICTING while `git merge-tree --write-tree` reported it clean at exit 0. The file is `.agents/issue-index.md`, which carries `merge=union` in `.gitattributes`; the forge does not apply that driver, so two appends read as a conflict there and merge without one here. This merge takes both appends and leaves every other row byte-for-byte unchanged. `f07f96e1c` (#1444) also edits `.agents/specs/expert-streaming.md` and `docs/USAGE.md`, neither of which this row touches. FOLLOWING_AGENTS_PROTOCOL Following-Agents-Protocol: true AI-Assisted: true Assisted-by: AGENT:claude-opus-5 [Claude Code]
localai-bot
pushed a commit
that referenced
this pull request
Aug 20, 2026
`origin/main` advanced again (#1444) during the gate run that picked up 9ecaf1b. Merged so the pull request stays MERGEABLE and its workflow runs against a base that carries the #1371 fix. The keyed record is verified BY KEY: 462 rows on `origin/main`, 466 merged, zero duplicated keys, zero lost keys on either side, every `origin/main` row and every branch row byte-for-byte equal in the result, `origin/main`'s order preserved as a subsequence, and the four branch-only rows (#926, #1290, #1416, #1431) byte-for-byte equal to the branch versions. FOLLOWING_AGENTS_PROTOCOL Following-Agents-Protocol: true AI-Assisted: true Assisted-by: AGENT:claude-opus-5 [claude-code]
localai-bot
pushed a commit
that referenced
this pull request
Aug 20, 2026
…ce was reddening this branch (#1395) FOLLOWING_AGENTS_PROTOCOL `build-test-cpu` and both `sanitize-cpu` jobs were red here on one test, `175 - test_nemotron_h_paged_forward`, and the defect was never in this branch. #1371 left `CPU_ATTN` as a name in a candidate list with no registrar, so the CPU inherited FlashAttention's `head_size % 8` constraint with no fallback. `9ecaf1bb3` (#1392) fixes that. This merge is taken so the gates judge the branch against the tree it will land on, and so `commit-trailers` and `commit-style` keep RUNNING instead of skipping on a branch that is behind. The merged commit is `f07f96e1c` (#1444) rather than `9ecaf1bb3`, because `origin/main` advanced by one documentation commit while this merge was being prepared. `9ecaf1bb3` is an ancestor of it, so the fix this merge is taken for is present either way. Merging the older SHA would have left the branch behind again and skipped the same two gates a second time. `.agents/issue-index.md` is the one keyed record both sides touch, and it was verified by key rather than trusted to the automatic merge: `origin/main` carries 462 rows and this tree carries 463, all 462 of main's rows are present byte-identical and in main's own order, the preamble is byte-identical, and the one added row is this branch's own `#1395`. Nothing is removed and no key is duplicated. No spec, record or product file is touched beyond what the merge itself requires. Following-Agents-Protocol: true AI-Assisted: true Assisted-by: AGENT:claude-opus-5 [claude-code]
localai-bot
pushed a commit
that referenced
this pull request
Aug 20, 2026
…ce was reddening this branch (#1313) FOLLOWING_AGENTS_PROTOCOL `build-test-cpu` and both `sanitize-cpu` jobs were red here on one test, `175 - test_nemotron_h_paged_forward`, and the defect was never in this branch. #1371 left `CPU_ATTN` as a name in a candidate list with no registrar, so the CPU inherited FlashAttention's `head_size % 8` constraint with no fallback. `9ecaf1bb3` (#1392) fixes that, and `f07f96e1c` (#1444) is the current `origin/main` that carries it. Merging the older SHA would have left this branch behind again, and `commit-trailers` and `commit-style` SKIP on a branch that is behind. A skip is not a green, which is the state this branch was already in. Four files are touched by both sides, and each was verified rather than trusted to the automatic merge. `.agents/issue-index.md` is the keyed record: main carries 462 rows and this tree carries 463, all 462 of main's rows are present byte-identical and in main's own order, the preamble is byte-identical, and the one added row is this branch's own `#1313`. Nothing is removed and no key is duplicated. For `docs/USAGE.md`, `src/vllm/v1/worker/gpu/runner.cpp` and `tests/CMakeLists.txt` the two sides edit disjoint regions, and that is measured rather than asserted: the added and removed content lines of `merged vs main` hash equal to this branch's own delta, and those of `merged vs branch` hash equal to main's own delta, for all three files. No file is touched beyond what the merge itself requires. Following-Agents-Protocol: true AI-Assisted: true Assisted-by: AGENT:claude-opus-5 [claude-code]
localai-bot
pushed a commit
that referenced
this pull request
Aug 20, 2026
…ce was reddening this branch (#1074) FOLLOWING_AGENTS_PROTOCOL `build-test-cpu` and both `sanitize-cpu` jobs were red here on one test, `175 - test_nemotron_h_paged_forward`, and the defect was never in this branch, which carries no product code at all. #1371 left `CPU_ATTN` as a name in a candidate list with no registrar, so the CPU inherited FlashAttention's `head_size % 8` constraint with no fallback. `9ecaf1bb3` (#1392) fixes that, and `f07f96e1c` (#1444) is the current `origin/main` that carries it. Merging the older SHA would have left this branch behind again, and `commit-trailers` and `commit-style` SKIP on a branch that is behind. Four files are touched by both sides, and each was verified rather than trusted to the automatic merge. `.agents/issue-index.md` is the keyed record: main carries 462 rows and this tree carries 464, all 462 of main's rows are present byte-identical and in main's own order, the preamble is byte-identical, and the two added rows are this branch's own `#1074` and `#1080`. Nothing is removed and no key is duplicated. `docs/BENCHMARKS.md`, `docs/FEATURES.md` and `docs/STATUS.md` are projections whose two sides edit different rows, and that is measured rather than asserted: the added and removed content lines of `merged vs main` hash equal to this branch's own delta, and those of `merged vs branch` hash equal to main's own delta, for all three files. No file is touched beyond what the merge itself requires. Following-Agents-Protocol: true AI-Assisted: true Assisted-by: AGENT:claude-opus-5 [claude-code]
localai-bot
pushed a commit
that referenced
this pull request
Aug 22, 2026
…ce was reddening this branch (#1074) FOLLOWING_AGENTS_PROTOCOL `build-test-cpu` and both `sanitize-cpu` jobs were red here on one test, `175 - test_nemotron_h_paged_forward`, and the defect was never in this branch, which carries no product code at all. #1371 left `CPU_ATTN` as a name in a candidate list with no registrar, so the CPU inherited FlashAttention's `head_size % 8` constraint with no fallback. `9ecaf1bb3` (#1392) fixes that, and `f07f96e1c` (#1444) is the current `origin/main` that carries it. Merging the older SHA would have left this branch behind again, and `commit-trailers` and `commit-style` SKIP on a branch that is behind. Four files are touched by both sides, and each was verified rather than trusted to the automatic merge. `.agents/issue-index.md` is the keyed record: main carries 462 rows and this tree carries 464, all 462 of main's rows are present byte-identical and in main's own order, the preamble is byte-identical, and the two added rows are this branch's own `#1074` and `#1080`. Nothing is removed and no key is duplicated. `docs/BENCHMARKS.md`, `docs/FEATURES.md` and `docs/STATUS.md` are projections whose two sides edit different rows, and that is measured rather than asserted: the added and removed content lines of `merged vs main` hash equal to this branch's own delta, and those of `merged vs branch` hash equal to main's own delta, for all three files. No file is touched beyond what the merge itself requires. Following-Agents-Protocol: true AI-Assisted: true Assisted-by: AGENT:claude-opus-5 [claude-code]
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
2 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.
docs/USAGE.mdstates that every decode figure in the Qwen3.8 reproduction section came from the W0e harness,benchmarks/expert_stream_device_w0e.cpp, rather than from thevllm-servercommand the section publishes. That disclosure is right for the W0e and W0f figures and wrong for one row: the 66.7 s/token streaming-off measurement of 16 August 2026 was itself a server run.The provenance is recorded and was checked at the source rather than inferred from the prose.
.agents/specs/expert-streaming.md:837describes the first run as--max-model-len 512,VT_GGUF_PREFAULT=0, server entry point, and:905records the 16 August run as the same vehicle and configuration with streaming off. So the one figure the section presents as measured on a different binary is the one figure measured on the same binary the reader is told to run.This scopes the sentence to the runs it is true of and names the exception, in
docs/USAGE.mdand in the spec paragraph that mirrors it. No figure changes and no new claim is added.Why it is a separate pull request
The correction was authored while #1211 was still open and was not pushed before that pull request merged as
b537a5344, so it did not land with the work it belongs to. The squash then removed the branch's merge base, and the branch cannot fast-forward ontomain. Cherry-picking the commit ontomainapplies cleanly.Gates
Cherry-pick rc 0, no conflicts.
check-agent-record.py,check-public-doc-tables.py,check-doc-checkpoint.pyandcheck-issue-index-append-only.pyall rc 0, run from this worktree's ownscripts/copy. The removed sentence is confirmed absent and the corrected form present, each grep run beside its counterpart so a null result cannot read as success. Two files, ten insertions, eight deletions, no product code.CI cannot give a verdict on this repository at present: every completed workflow run returns
cancelled(#1285). That isREMOTE_UNVERIFIEDand is not a pass.Closes #1447.
FOLLOWING_AGENTS_PROTOCOL
Following-Agents-Protocol: true
AI-Assisted: true
Assisted-by: AGENT:claude-opus-5 [Claude Code]