No. 018 · 27 August 2026 · 3 min← No. 017 · No. 019 →

Correction

Issue 018 published late (~08:30 UTC) after the morning desk run did not land the file; the collection window stays 24h to 06:00 UTC 27 Aug.

SIGNALNORMAL|SHIPPED 2|SKIPPED 6Cursor raises included Grok / first-party usageRETIRING MAI-Code-1-Flash (Copilot) · 14d

The one thing

vLLM 0.28.0 ships Model Runner V2 defaults and breaks bitsandbytes-in-tree; Claude Code latest is 2.1.247 while stable stays 2.1.231.

vLLM v0.28.0 (26 Aug) raises max_num_batched_tokens from 8192 to 16384, enables Mamba prefix caching by default, and matures Model Runner V2 (E/P/D disaggregation, tiered KV offload including disk, multi-layer MTP KV).¹ Breaking: bitsandbytes moves out-of-tree; Transformers floor is 5.15.0; calculate_kv_scales and override_attention_dtype are removed.¹

npm @anthropic-ai/claude-code: stable still 2.1.231; latest/next → 2.1.247.²³ 2.1.247 adds SendFeedback drafts and fixes /compact / --agent summarization under the wrong system prompt.³ Channel policy from #015 stands.

VERDICT · BREAKING — vLLM: checklist bnb plugin + Transformers ≥5.15 before upgrade. Claude Code: latest ≥2.1.247 for those fixes; stable remains 2.1.231.

Deeper — vLLM 0.28.0 defaults and breaks

Defaults: batched tokens 16384; Mamba prefix cache on.¹
Breaks: bitsandbytes plugin; Transformers ≥5.15.0; drop two deprecated knobs.¹
Image: vllm/vllm-openai:v0.28.0 (cu130) plus cu129/ROCm/CPU/XPU tags.¹

Shipped

Breaking

vLLM 0.28.0 · Runner V2, tiered KV offload, bitsandbytes out-of-tree

Major engine cut (26 Aug): Runner V2, tiered KV with disk offload, Rust/gRPC multimodal path, new models (Muse Glimmer, Ling 3.0 Flash, Dots3 NOTE).¹ Batch token default doubles to 16384. bitsandbytes is a plugin; pin Transformers ≥5.15.0.¹

MATTERS TO · anyone running vLLM servers or pinning CUDA/ROCm serving images

nofeed.dev/issues/2026-08-27/vllm-0-28-0/

Use it

Codex 0.150.0 · @task mentions, Interrupt hooks, /copy picker

rust-v0.150.0 (26 Aug; 0.150.1 same day): @ mentions across Codex tasks; /copy picker; auto titles; Interrupt hooks for commands/MCP on turn cancel; permission-mode shortcuts; Vim . repeat.⁶⁷

MATTERS TO · terminal Codex users and teams wiring interrupt-time automation

nofeed.dev/issues/2026-08-27/codex-0-150-0/

Promised, not shipped
OpenAI Private Safety Processing / frontier ZDR — September white paper (issue 013) · Origin agent-native source-control — early beta ongoing (issue 011) · Claude Code /design skill — research preview · MAI-Code-1-Flash off Copilot — 10 Sept 2026 · Grok Bot enterprise — waitlist · OpenAI Ultrafast — expands with capacity · Ox Alpha free window — ~through 27 Aug (ends today) · GPT-5.6 Sol promo pricing — at least through 21 Nov 2026 · Plus Codex/Work 5h limit — live from 25 Aug; Pro stays off for upcoming months · GitHub Copilot global model policy enforcement — rolling through 1 Sept 2026

The conversation

The claim — including ours

vLLM 0.28 is a real serving break; Claude Code still has no native AGENTS.md; Cursor permanently raised included Grok usage.¹⁸⁹

From the floor

vLLM v0.28.0 release notesPrimary, 26 Aug
Defaults up; bitsandbytes out-of-tree; Transformers ≥5.15.¹
@trq212 (Claude Code) via X digestVendor, digest 26 Aug 05:30 (~24.5h old at window close)
Easier AGENTS.md / prompt mods coming; per-model prompts stay; interim @AGENTS.md from CLAUDE.md.⁸
@leerob (Cursor) via X digestVendor, same digest
Included Grok usage raised permanently across first-party Cursor models.⁹

Our take

Upgrade vLLM only after the bnb/Transformers checklist. AGENTS.md native support is still unshipped. Re-check Cursor quotas if Grok is default.

GitHub releases: 28 repos, sweep_complete true, authenticated, 19 in-window. Vendor feeds: 12 sources, sweep_complete true, 10 in-window (Actions critical 2.8h resolved; Actions+PR 1.5h resolved; Billing minor open). X accounts digest (~101 roster) 05:30 UTC 26 Aug (~24.5h old at window close) — colour only. X keyword same age. Reddit pulse (6 subs, 1 post). HN in-window: GLM-5.3-Flash, Ox Alpha, GitHub disruption. arXiv not swept. Morning cron missed; recovered ~08:30 UTC.

Skip this

GitHub Actions critical 2.8h; Actions+PR minor 1.5h; Billing minor open. Resolved or open weather with duration — not a pin. Billing still open at sweep; watch status if invoices fail.

GitHub Copilot global model policy GA (rollout through 1 Sept). Enterprise admin defaulting change. Open-weight and data-retention models stay off by default. Promised, not a day-of ship for most readers.

OpenAI Admin plugin for ChatGPT Work & Codex; Hugging Face incident post. Admin surface consolidation. HF incident page timed out on fetch — not printed as fact.

Transformers 5.16.0/5.16.1 (DTensor TP, Fuyu image_patch_indices drop, NVFP4). Real breaks for TP/Fuyu stacks, but secondary to the vLLM floor note already in One thing.

openai-python 3.4.0/3.5.0; anthropic-sdk-python 1.1.0; adk-python 2.8.0; ollama 0.33.1; MCP Python 2.0.1; cline 4.1.16/cli 3.0.60; AI SDK patches. Optional call IDs, thinking display beta, ADK live/data-agent tools, Qwen3.8 Flash Next on MLX, FastMCP import pointer, Cline hub polish — no default-stack pin today.

GLM-5.3-Flash / Qwen3.8-Flash-Next HN spikes; Nvidia–Hugging Face $13B rumour. Model drops without a verified serving pin for this audience. Acquisition item is Business Insider talk, not a closed deal we opened.

Everything we saw

55 candidates scanned · 8 used in this issue — the rest, with the reason each one was left out

The receipt. Ranking is only trustworthy if the discarded pile is visible, so here it is: everything the collectors surfaced in the window, with its signal and what we did with it.

Item	Source	Signal	Call
vLLM v0.28.0	github releases	22.8h	led the issue
npm @anthropic-ai/claude-code dist-tags	npm registry	stable 2.1.231 / latest 2.1.247	one thing
Claude Code v2.1.247	github releases	9.4h	one thing
Transformers v5.16.0	github releases	20.0h	skipped secondary
Transformers v5.16.1	github releases	17.7h	skipped secondary
Codex rust-v0.150.0	github releases	12.9h	shipped
Codex rust-v0.150.1	github releases	6.6h	shipped patch
X accounts digest 26 Aug 05:30	digests	~24.5h old; 101 accounts	conversation colour
GitHub global model policy GA	github changelog	10.4h	skipped / promised
GitHub Actions critical incident	github status	2.8h critical resolved	skipped weather
openai-python v3.5.0	github releases	7.6h	skipped
anthropic-sdk-python v1.1.0	github releases	15.3h	skipped
adk-python v2.8.0	github releases	9.1h	skipped
ollama v0.33.1	github releases	14.4h	skipped
MCP Python SDK v2.0.1	github releases	21.8h	skipped
HN — GLM-5.3-Flash	news.ycombinator.com	1011 pts / 507 cmt	skipped model drop
HN — Ox Alpha / Z.ai weights	news.ycombinator.com	425 pts	skipped
Reddit pulse — 6 subs, 1 post	reddit pulse	layer ran	layer ran
OpenAI — Hugging Face incident post	openai.com	crawl timeout	skipped unverified

EOF

End of feed. That is everything from the window worth your time.
Explore more news, analysis and practical guides on NO FEED.

Still developingClaude Code stable lag (015/017/018). AGENTS.md native still unshipped. Origin beta (011). Ultrafast capacity (009). MAI-Code-1-Flash off Copilot 10 Sept. ZDR white paper September. Ox Alpha free ends ~27 Aug. Sol promo ≥21 Nov. Plus Codex/Work 5h from 25 Aug. Copilot global model policy through 1 Sept.

Sources — each with the date it was read01 github — vLLM v0.28.0 (2026-08-27) · 02 npm — @anthropic-ai/claude-code dist-tags (2026-08-27) · 03 github — claude-code v2.1.247 (2026-08-27) · 06 github — codex rust-v0.150.0 (2026-08-27) · 07 github — codex rust-v0.150.1 (2026-08-27) · 08 X — @trq212 on AGENTS.md / Claude Code hackability (2026-08-27) · 09 X — @leerob Cursor included Grok usage raise (2026-08-27)