Lesson 2 · Boundary and attributionCourse indexPT-BR
CLIProxyAPI benchmark · Boundary and attribution

Qualify six lanes

A registered endpoint does not prove every model works through every protocol or harness.

Primary local sources: BENCHMARK-MEASUREMENT-PLAN.md, model-context-matrix.csv, and planned-run-manifest.json.
BENCHMARK-MEASUREMENT-PLAN.md · model-context-matrix.csv · planned-run-manifest.json

The idea in plain language

The candidate grid has six text lanes: direct Chat, Responses, and Messages; Warp Chat; Codex Responses; and Claude Code Messages. Each lane × model pair enters only after a schema, authentication, and identity canary. Images remain outside this ranking.

Running example: send the same sealed parcel — nonce, checksum, and three needles — through all six lanes. If the harness changes the wrapping, you are measuring the lane too. The analogy breaks because context, tools, and thinking carry semantics, not size alone.

An unauthenticated POST returned 401 rather than 404 on registered endpoints. That proves route registration, not per-model compatibility. Codex and Claude Code must still prove authentication and envelope behavior; the Warp sidecar transforms Chat and emulates Responses over Chat.

Verifiable excerpt

direct_chat · direct_responses · direct_messages\nwarp_chat · codex_responses · claude_code_messages\nqualification = 256-token schema + auth + identity canary

outputs/BENCHMARK-MEASUREMENT-PLAN.md · outputs/model-context-matrix.csv · outputs/planned-run-manifest.json

The path in one picture

UI / CLIobserved outputharnessWarp · Codex · Claudecompatibilitysidecar / envelopeCLIProxyAPI127.0.0.1:8317providerdirect Chat · Responses · MessagesPRIMARY BOUNDARY
Running example: the same nonce + checksum crosses every lane; the failure belongs to the smallest boundary where it reproduces.

Retrieval check

What does a 401 on the endpoint prove?