# Benchmark data and API

Operated by the Busbar team; Busbar is also an entrant. Gateway overhead against a local mock is not end-to-end model latency.  [Methodology and limitations](https://onthebench.ai/gateways/method) .

 

Published collection: dataset-c7c493902ee4473d65d1886e06f885af4adac2f4f5a4ea5fc70f2efa5f230c2f. Generated 2026-11-14T11:09:24.858Z; measurements 2026-11-10T17:39:19Z through 2026-11-11T03:17:59Z. Metadata refresh is separate from measurement time. This is a collection of independent runs, not a claim of one complete simultaneous field run.

 

[API discovery](https://onthebench.ai/api/v1)  ·  [Latest JSON](https://onthebench.ai/api/v1/benchmarks/latest)  ·  [Markdown](https://onthebench.ai/gateways/index.md)  ·  [Compact CSV](https://onthebench.ai/gateways/data.csv?compact=1)  ·  [OpenAPI](https://onthebench.ai/openapi.json)  ·  [JSON schemas](https://onthebench.ai/api/v1/schema)  ·  [Historical runs](https://onthebench.ai/gateways/runs)

 

[Agent discovery (llms.txt)](https://onthebench.ai/llms.txt)  ·  [Full agent guide](https://onthebench.ai/llms-full.txt)  ·  [Version 1 measurement schema](https://onthebench.ai/api/v1/schema/measurement.json)

 

## Per-gateway exports

- agentgateway:  [Markdown](https://onthebench.ai/gateways/agentgateway.md)  ·  [Compact CSV](https://onthebench.ai/gateways/data.csv?gateway=agentgateway&compact=1)
- AISIX (api7):  [Markdown](https://onthebench.ai/gateways/aisix.md)  ·  [Compact CSV](https://onthebench.ai/gateways/data.csv?gateway=aisix&compact=1)
- APISIX:  [Markdown](https://onthebench.ai/gateways/apisix.md)  ·  [Compact CSV](https://onthebench.ai/gateways/data.csv?gateway=apisix&compact=1)
- Bifrost:  [Markdown](https://onthebench.ai/gateways/bifrost.md)  ·  [Compact CSV](https://onthebench.ai/gateways/data.csv?gateway=bifrost&compact=1)
- Busbar:  [Markdown](https://onthebench.ai/gateways/busbar.md)  ·  [Compact CSV](https://onthebench.ai/gateways/data.csv?gateway=busbar&compact=1)
- GoModel:  [Markdown](https://onthebench.ai/gateways/gomodel.md)  ·  [Compact CSV](https://onthebench.ai/gateways/data.csv?gateway=gomodel&compact=1)
- Helicone:  [Markdown](https://onthebench.ai/gateways/helicone.md)  ·  [Compact CSV](https://onthebench.ai/gateways/data.csv?gateway=helicone&compact=1)
- Higress:  [Markdown](https://onthebench.ai/gateways/higress.md)  ·  [Compact CSV](https://onthebench.ai/gateways/data.csv?gateway=higress&compact=1)
- Kong:  [Markdown](https://onthebench.ai/gateways/kong.md)  ·  [Compact CSV](https://onthebench.ai/gateways/data.csv?gateway=kong&compact=1)
- LiteLLM · Python:  [Markdown](https://onthebench.ai/gateways/litellm-python.md)  ·  [Compact CSV](https://onthebench.ai/gateways/data.csv?gateway=litellm-python&compact=1)
- LiteLLM · Rust:  [Markdown](https://onthebench.ai/gateways/litellm-rust.md)  ·  [Compact CSV](https://onthebench.ai/gateways/data.csv?gateway=litellm-rust&compact=1)
- One-API:  [Markdown](https://onthebench.ai/gateways/one-api.md)  ·  [Compact CSV](https://onthebench.ai/gateways/data.csv?gateway=one-api&compact=1)
- Plano:  [Markdown](https://onthebench.ai/gateways/plano.md)  ·  [Compact CSV](https://onthebench.ai/gateways/data.csv?gateway=plano&compact=1)
- Portkey:  [Markdown](https://onthebench.ai/gateways/portkey.md)  ·  [Compact CSV](https://onthebench.ai/gateways/data.csv?gateway=portkey&compact=1)
- TensorZero:  [Markdown](https://onthebench.ai/gateways/tensorzero.md)  ·  [Compact CSV](https://onthebench.ai/gateways/data.csv?gateway=tensorzero&compact=1)

## Compact results and freshness

Start with  [the compact summary](https://onthebench.ai/api/v1/summary)  (one protocol pair, core metrics and provenance). Filter with gateway, ingress and egress. Full collections may exceed retrieval-tool limits; follow paginated measurement links for details. Each compact result and comparison measurement includes measured_at, published_at, generated_at, age, age_days and age_as_of. Age is measured at collection generation, not at request time. published_at is recorded for new releases when publishing begins; it remains null for local builds and older releases without that record.

current_metadata contains repository stars and their stars_as_of date. Legacy gateway.stars fields refer to that same metadata snapshot, not the benchmark date. Stars at measurement time were not recorded and are not inferred.

## Context-efficient retrieval

Use `/api/v1/measurements?compact=1&limit=10` to receive definitions once, keyed by definition_id. Compact summary uses the same dictionary pattern. Prefer each gateway’s Markdown and `/gateways/data.csv?gateway=ID&compact=1` links; the CSV references  [the definition catalog](https://onthebench.ai/api/v1/definitions) . Full CSV and full JSON are bulk exports.  [The paginated run catalog](https://onthebench.ai/api/v1/runs)  supports exact gateway IDs and families such as `family=busbar`.

## Retrieval

All resources are public and read-only. HTML contains measurements without JavaScript. Explicit Markdown and JSON links avoid content negotiation. Query endpoints return 400 for invalid filters and 404 for unknown entities.

 

```
GET /api/v1/summary?ingress=openai&egress=openai
GET /api/v1/gateways
GET /api/v1/gateways/busbar
GET /api/v1/measurements?gateway=busbar&ingress=openai&egress=openai&limit=50
GET /api/v1/compare?a=busbar&b=bifrost&ingress=openai&egress=openai&metric=added_latency_p99_us
```

 

Measurements support gateway, ingress, egress, metric, limit (1–100), and offset (0 or greater). Responses are capped at 1 MiB of uncompressed UTF-8 JSON, so a page may contain fewer than limit rows. Follow the returned next URL until null; do not infer completion from row count. An individual oversized record returns 413 with narrowing guidance. Comparisons require two distinct published gateway IDs and a recorded metric. The default protocol pair is OpenAI → OpenAI and the default metric is added_latency_p99_us. Available gateway IDs and metrics are listed by the API.

 

## Compare and cite using HTTP only

 
- Discover gateway IDs with  [the catalog](https://onthebench.ai/api/v1/gateways)  and choose an explicit ingress/upstream pair. The API calls the upstream protocol `egress`. 
- Retrieve  [OpenAI → Anthropic measurements](https://onthebench.ai/api/v1/measurements?ingress=openai&egress=anthropic&limit=100) ; follow `next` until it is null. Unavailable values are not zero. 
- Request  [Busbar vs LiteLLM on that pair](https://onthebench.ai/api/v1/compare?a=busbar&b=litellm-python&ingress=openai&egress=anthropic&metric=added_latency_p99_us)  and inspect `comparability.status`. Only `comparable` supports a ratio; `incomparable` and `insufficient_evidence` include reasons. Matching protocol names alone is not enough. 
- Cite each record’s `provenance.run_url`, tested `gateway_build`, `measured_at`, protocol pair and metric definition. `provenance.raw_result_url` and `raw_sha256` identify the exact source bytes. Include  [the methodology](https://onthebench.ai/gateways/method)  and operator disclosure. 

The default dashboard is not a like-for-like ranking. Its representative protocol cell can differ between products. The API returns per-cell evidence and conservative pairwise comparisons; it does not publish an unconditional fastest-gateway ranking. These are independent gateway runs against a local mock, not end-to-end model latency.

 

## Interpretation

Ratios require an identical protocol pair, hardware, architecture, metric definition, clean harness commit, recorded mock digest, and metric-specific workload. Missing evidence never matches missing evidence. These are observed gateway measurements against a local mock, not a claim about model latency or statistical significance.

Every metric carries its original definition, state, units in its field name, and provenance when recorded. Missing evidence remains null. A measured zero is retained; a below-resolution reading has a null value and an explicit state. Throughput readings carry their tail-latency bound, operating concurrency, and lower-bound status. Compare conditions before ranking.

 

## Immutable evidence

Run IDs include the SHA-256 of their exact raw JSON bytes. Raw files contain the complete samples; dashboard drawing series are thinned. Corrected raw content receives a different ID. The latest collection points to individual runs and retains their actual dates. A collection ID identifies its generated evidence representation; it is distinct from a per-gateway run ID.

 

## Dates and licensing

published_at is the release timestamp recorded when publishing begins, not the original publication date of historical measurements; generated_at is collection build time; measured_at belongs to the instrument; stars_as_of belongs to metadata. An original first-publication timestamp is not inferred. Original measurements and dataset compilation are licensed under  [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) . Credit On the Bench — Busbar team, link the source and license, and indicate modifications.  [Dataset license and scope](https://onthebench.ai/gateways/license) ; software and third-party materials retain their own terms.
