tenjin.blog

https://tenjin.blog/api/read/arxiv-ml/paper-trail-evaluation-claims-expire
down

Shares a payout wallet with 3 other endpoints — its adoption is attributed from that shared wallet's traffic, not earned exclusively.

Paper Trail: Evaluation Claims Expire. Two July papers argue that benchmark scores need scope and expiration metadata; WorkSurface-Bench, a synthetic-user benchmark, a coding-agent user study, and a long-context aggregation method show what gets lost when a score is treated as portable evidence.

37
reliability

Reliability & health

Uptime and response time from our own unpaid probes, plus the reliability score built from them.

Uptime

56.9% avg
0% 50% 100% Sat 9 AMSun 5 AMMon 12 AMWed 5 PM

Response time

p50 118ms · p95 184ms
0ms 100ms 200ms Sat 9 AMSun 5 AMMon 12 AMWed 5 PM
Data table
Reliability history per time bucket over the last 7 days (hourly).
Bucket Uptime Probes Response time
8/15/2026, 9:00:00 AM 100% 2 170ms
8/15/2026, 10:00:00 AM 100% 1 132ms
8/15/2026, 12:00:00 PM 100% 1 118ms
8/15/2026, 1:00:00 PM 100% 1 103ms
8/15/2026, 2:00:00 PM 100% 1 123ms
8/15/2026, 3:00:00 PM 100% 1 128ms
8/15/2026, 4:00:00 PM 100% 1 193ms
8/15/2026, 5:00:00 PM 100% 1 105ms
8/15/2026, 7:00:00 PM 100% 1 101ms
8/15/2026, 8:00:00 PM 100% 1 118ms
8/15/2026, 9:00:00 PM 100% 1 111ms
8/15/2026, 10:00:00 PM 100% 1 112ms
8/15/2026, 11:00:00 PM 100% 1 174ms
8/16/2026, 12:00:00 AM 100% 1 99ms
8/16/2026, 2:00:00 AM 100% 1 154ms
8/16/2026, 3:00:00 AM 100% 1 104ms
8/16/2026, 4:00:00 AM 100% 1 106ms
8/16/2026, 5:00:00 AM 100% 1 144ms
8/16/2026, 6:00:00 AM 100% 1 169ms
8/16/2026, 7:00:00 AM 100% 1 130ms
8/16/2026, 9:00:00 AM 100% 1 101ms
8/16/2026, 10:00:00 AM 100% 1 102ms
8/16/2026, 11:00:00 AM 100% 1 95ms
8/16/2026, 12:00:00 PM 100% 1 102ms
8/16/2026, 1:00:00 PM 100% 1 136ms
8/16/2026, 2:00:00 PM 100% 1 94ms
8/16/2026, 4:00:00 PM 100% 1 112ms
8/16/2026, 5:00:00 PM 100% 1 122ms
8/16/2026, 6:00:00 PM 100% 1 127ms
8/16/2026, 7:00:00 PM 0% 1
8/16/2026, 8:00:00 PM 0% 1
8/16/2026, 9:00:00 PM 0% 1
8/16/2026, 11:00:00 PM 0% 1
8/17/2026, 12:00:00 AM 0% 1
8/17/2026, 1:00:00 AM 0% 1
8/17/2026, 2:00:00 AM 0% 1
8/17/2026, 3:00:00 AM 0% 1
8/17/2026, 4:00:00 AM 0% 1
8/17/2026, 6:00:00 AM 0% 1
8/17/2026, 7:00:00 AM 0% 1
8/17/2026, 8:00:00 AM 0% 1
8/17/2026, 9:00:00 AM 0% 1
8/17/2026, 10:00:00 AM 0% 1
8/17/2026, 11:00:00 AM 0% 1
8/17/2026, 1:00:00 PM 0% 1
8/17/2026, 2:00:00 PM 0% 1
8/17/2026, 3:00:00 PM 0% 1
8/17/2026, 4:00:00 PM 0% 1
8/17/2026, 5:00:00 PM 0% 1
8/18/2026, 5:00:00 PM 0% 1
8/19/2026, 5:00:00 PM 0% 1
37
Composite reliability — the weighted average of the dimensions below, on a 0–100 scale.
Availability
20

How often it was reachable when we probed it (weight 35%).

Latency
100

How quickly it answers when it is up — fast scores high (weight 20%).

Security
100

Whether it serves over working TLS / https (weight 15%).

Transparency
36

Whether the operator disclosed what it does — description, schema, identity (weight 15%).

Activity
1

Observed on-chain payments to this endpoint's wallet — transactions, volume, and distinct paying wallets over the last month (weight 15%; not yet measured when we have not seen its wallet settle).

Confidence 10% — how much probe evidence backs this score. A thin history is flagged, not hidden.

See the scoring methodology for exactly how each number is computed.

Payment

Currencies
USDC
Networks
base
Facilitator
unknown
SDK
unknown

Operator

Company
inferred — unknown
Jurisdiction
Registrar
Namecheap
Domain registered
2026-04-21
Hosting network
VERCEL-01
Hosting country

Operator context is derived from public RDAP/TLS records and unverified. Company and country are only available for endpoints that publish them (an OV/EV TLS certificate or a non-redacted domain record); most endpoints are CDN-fronted and disclose neither.

On-chain activity (base)

1
payments (30d)
0.1
USD volume (30d)
1
paying wallets (30d)

Payments per day

last payment 7/30/2026
0 3 5 Jul 30
Data table
Settled payments per day.
Day Payments Volume (USD) Paying wallets
7/30/2026 1 0.1 1

This payout wallet is shared: figures cover all 4 endpoints paid at it, since payments cannot be attributed to one of them. Only payments settled on chains and facilitators we index.