Columns
| Column | Type | Size | Nulls | Auto | Default | Children | Parents | Comments | |||
|---|---|---|---|---|---|---|---|---|---|---|---|
| id | serial | 10 | √ | nextval('quality_scores_id_seq'::regclass) |
|
|
{“what”:“Unique integer row identifier.”,“why”:“Provides a stable primary key for joins.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:59.517864481+00:00”} |
||||
| provider_id | text | 2147483647 | null |
|
|
{“what”:“Provider name such as ‘openai’ or ‘anthropic’.”,“why”:“Separates scores by vendor.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:59.534933691+00:00”} |
|||||
| model_id | text | 2147483647 | null |
|
|
{“what”:“Model name such as ‘gpt-4’ or ‘claude-3’.”,“why”:“Identifies which model the scores belong to.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:59.555376701+00:00”} |
|||||
| tool_use | float8 | 17,17 | √ | null |
|
|
{“what”:“Score for tool-use ability; higher is better.”,“why”:“Without it, tool-use performance cannot be compared.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:59.574202811+00:00”} |
||||
| reasoning | float8 | 17,17 | √ | null |
|
|
{“what”:“Score for reasoning ability; higher is better.”,“why”:“Without it, reasoning performance cannot be compared.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:59.589797538+00:00”} |
||||
| code_gen | float8 | 17,17 | √ | null |
|
|
{“what”:“Score for code generation quality; higher is better.”,“why”:“Without it, code generation performance cannot be compared.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:59.607105515+00:00”} |
||||
| instruction_following | float8 | 17,17 | √ | null |
|
|
{“what”:“Score for instruction following; higher is better.”,“why”:“Without it, instruction following performance cannot be compared.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:59.631967446+00:00”} |
||||
| composite | float8 | 17,17 | √ | null |
|
|
{“what”:“Aggregate score combining other metrics.”,“why”:“Provides a single number for quick model ranking.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:59.655559271+00:00”} |
||||
| perplexity | float8 | 17,17 | √ | null |
|
|
{“what”:“Perplexity value; lower is better.”,“why”:“Indicates model confidence; missing if not computed.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:59.685156156+00:00”} |
||||
| sample_count | int4 | 10 | √ | null |
|
|
{“what”:“Number of samples used for the scores.”,“why”:“Indicates reliability; missing if unknown.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:59.702559661+00:00”} |
||||
| source | text | 2147483647 | √ | 'openrhythm'::text |
|
|
{“what”:“Origin of the scores, default ‘openrhythm’.”,“why”:“Tracks which evaluation pipeline produced the data.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:59.717868969+00:00”} |
||||
| measured_at | timestamptz | 35,6 | now() |
|
|
{“what”:“UTC timestamp of measurement.”,“why”:“Enables time-based analysis and freshness checks.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:59.735950050+00:00”} |
Indexes
| Constraint Name | Type | Sort | Column(s) |
|---|---|---|---|
| quality_scores_pkey | Primary key | Asc | id |
| idx_quality_lookup | Performance | Asc/Asc/Desc | provider_id + model_id + measured_at |