财经媒体的独立表态：TechStock²《…Its 445× Cost Claim Is Still Self-Tested》（2026-09-17）：*"…leaves the biggest performance numbers dependent on TypeSafe…
TechStock2 article noting that TypeSafe's headline Jev cost and performance claims, such as 445x cheaper, are self-tested.
This summary is based on how curated lists describe the project; its README was not available to review.
Overview
This is a TechStock2 (ts2.tech) article from 17 September 2026 covering TypeSafe AI's funding round for Jev. According to its title and the list entry, it points out that Jev's biggest performance numbers, including a claim of being 445 times cheaper, come from TypeSafe's own testing rather than independent evaluation.
The directory has no further text from the article. It is useful as an example of press coverage that questions vendor-reported benchmarks and as context for the many community benchmarks that try to measure Jev independently.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Other | Are Jev's headline cost and performance claims independently verified?(inferred) | — |
How it works
- Report on the funding announcement.
- Note which performance figures come from TypeSafe itself.
- Highlight the lack of independent testing.
Strengths
- Provides a skeptical view of vendor-reported numbers.
Caveats
- Only the title and URL are captured here.
- News coverage, not a technical evaluation.
At a glance
- Type
- Article
- Status
- discussion
- Has code
- no
- Has tests
- no
Use cases and integrations
How curated lists describe it
Independent statement from financial media: \.
wh000wh000/awesome-jev-live
Frequently asked questions
What is the article about?
TypeSafe AI's funding for Jev and the fact that its headline cost claim is self-tested.
Does it include new benchmark results?
Not as far as the captured title indicates; it comments on TypeSafe's own figures.
Where can I read it?
On ts2.tech at the linked URL.
Related projects
Chinese-run aggregator of X long posts and demo videos about Jev, keeping original links and authors rather than rehosting.
Serviceproductioncellcog.ai page tracking Jev that keeps layered, dated records of what is claimed, conceded, measured and found.
ArticlediscussionSimon Willison's blog post on what changes when a model's output is a typed decision instead of prose.
ArticlediscussionSubstack essay arguing cheap calibrated decisions will shift where AI inference spending goes.
ArticlediscussionTypeSafe's manifesto arguing for machine-native intelligence built for software rather than conversation.
ArticlediscussionArcturus Labs analysis arguing OpenAI is best placed to fast-follow Jev's typed-decision shape, with a large HN debate on moats.
Articlediscussion