mcpbench GitHub

server/stateless-resources

openai/gpt-5.6-sol:medium 2026-07-28 docs: none score 0% task-failure musts 0/3 · shoulds 0/0 · mays 0/0 works no · conformant no

Checkpoints

MUST (any failure zeroes the score — 0/3 passed)

Checkpoint Description Detail
read-works resources/read of the declared URI returns its text status 404, expected 200
list-works resources/list returns the resource listing status 404, expected 200
unknown-uri-errors resources/read of a nonexistent URI returns the invalid-params error (-32602) on a 200 status 404, expected 200

Usage & cost

Nominal cost
$0.212
Input tokens
10,019
Output tokens
5,217
Cache read
10,752
Duration
119s
Timestamp
2026-07-27 17:23:19Z
Bench version
0.3.0

Cost is nominal (public API pricing): token usage × published rates; actual marginal cost is subscription-covered.