TrackLLM: Are the LLM APIs you rely on stable?
Latest changes how far each endpoint moved from its baseline All 191 changes → 11 Sep qwen3-max-thinking served by Alibaba Cloud Probe-prompt outputs shifted (total variation 0.20) B3IT 11 Sep glm-5.3 served by Sail Research (fp8) Probe-prompt outputs shifted (total variation…
- ▪Latest changes how far each endpoint moved from its baseline All 191 changes → 11 Sep qwen3-max-thinking served by Alibaba Cloud Probe-prompt outputs shifted (total variation 0.20) B3IT 11 Sep glm-5.3 served by Sail Research (fp8) Probe-pro
Hacker News (AI / LLM) files mainly under ai. We currently carry 4,752 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | Trackllm |
| Canonical URL | https://www.trackllm.net/ |
| Publication time | Mon, 14 Sep 2026 06:38:39 +0000 |
| Retrieval time | 2026-09-14T06:46:51.193Z |
| Last seen | 2026-09-14T06:46:51.193Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | QYYGKUz3D33S · 1 stories |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
Latest changes how far each endpoint moved from its baseline All 191 changes → 11 Sep qwen3-max-thinking served by Alibaba Cloud Probe-prompt outputs shifted (total variation 0.20) B3IT 11 Sep glm-5.3 served by Sail Research (fp8) Probe-prompt outputs shifted (total variation 0.23) B3IT 10 Sep glm-4.7-flash served by Venice AI (fp8) Probe-prompt outputs shifted (total variation 0.38) B3IT 9 Sep claude-sonnet-4.5 served by Claude on AWS Probe-prompt outputs shifted (total variation 0.21) B3IT 7 Sep deepseek-v4-flash-0731 served by Mancer (fp8) Probe-prompt outputs shifted (total variation 0.36) B3IT 6 Sep qwen3.6-max-preview served by Alibaba Cloud Logprob averages moved 3.00 nats from baseline LT 5 Sep gemini-3.1-pro-preview served by Google AI Studio (flex) Probe-prompt outputs shifted…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Trackllm.
Discussion
0 commentsMore from Trackllm
%3Aquality(75)%2Fhttps%253A%252F%252Fassets.lareviewofbooks.org%252Fuploads%252F201311when-god-spoke-greek-the-septuagint-and-the-making-of-the-christian-bible_6575_5001.jpg&w=320)