Models
The leaderboard is the catalog, sorted by what people actually use
300+ models are routable. The tables below rank the ones moving real traffic, with the numbers straight from the routing layer: tokens, requests, median latency, context, and price. No editorial picks, no vendor claims.
- Helios 5
- Meridian Max 4.1
- Tycho 2.5 Deep
- Foghorn 4 Coastal
- Northbeam V3.1
- Vela 3 Max
- Aurora Large 3
- Kestrel 4
- Beacon A
- Argosy K2
- Tidal 1.1 Pro
- Maelstrom 3
- Echo large-v3
- Siren Multilingual v2
03 / Leaderboard
What the world actually routes, refreshed weekly
Ranked by routed tokens over the last 30 days across the gateway. Snapshot: Week 39, 2026. Every figure comes from real traffic, not vendor claims.
| # | Model | Provider | Tokens, 30d | Requests, 30d | Share | P50 latency | Context | Price in/out | Free |
|---|---|---|---|---|---|---|---|---|---|
| 01 | Helios 5 | Solstice Labs | 412.8B | 148.2M | 320 ms | 400k | $1.25 / $10.00 | - | |
| 02 | Meridian Pro 4.5 | Meridian AI | 356.2B | 121.7M | 280 ms | 200k | $3.00 / $15.00 | - | |
| 03 | Tycho 2.5 Swift | Tycho Systems | 289.6B | 174.3M | 190 ms | 1M | $0.30 / $2.50 | - | |
| 04 | Foghorn 4 Coastal free | Foghorn AI | 201.4B | 96.8M | 240 ms | 1M | $0.22 / $0.85 | yes | |
| 05 | Northbeam V3.1 free | Northbeam | 168.9B | 71.2M | 410 ms | 128k | $0.56 / $1.68 | yes | |
| 06 | Helios 5 Mini | Solstice Labs | 122.7B | 139.5M | 210 ms | 400k | $0.25 / $2.00 | - | |
| 07 | Vela 3 235B free | Vela Labs | 98.3B | 48.9M | 330 ms | 256k | $0.20 / $0.60 | yes | |
| 08 | Meridian Max 4.1 | Meridian AI | 87.5B | 19.4M | 520 ms | 200k | $15.00 / $75.00 | - | |
| 09 | Tycho 2.5 Deep | Tycho Systems | 74.2B | 26.8M | 460 ms | 1M | $1.25 / $10.00 | - | |
| 10 | Aurora Large 3 | Aurora Compute | 52.8B | 22.1M | 290 ms | 256k | $2.00 / $6.00 | - | |
| 11 | Kestrel 4 | Kestrel AI | 43.6B | 14.6M | 350 ms | 256k | $3.00 / $15.00 | - | |
| 12 | Pelagic 70B free | Pelagic AI | 38.1B | 51.7M | 200 ms | 128k | $0.12 / $0.30 | yes | |
| 13 | Beacon A | Lighthouse Labs | 24.9B | 9.3M | 260 ms | 256k | $2.50 / $10.00 | - | |
| 14 | Argosy K2 | Argosy AI | 18.6B | 7.8M | 380 ms | 256k | $0.60 / $2.50 | - |
Multimodal board
Non-text models, ranked by routed requests over the same 30 day window.
| # | Model | Provider | Modality | Requests, 30d | Unit price |
|---|---|---|---|---|---|
| 01 | Echo large-v3 | Meridian AI | speech-to-text | 22.6M | $0.006 / min |
| 02 | Siren Multilingual v2 | Kestrel AI | text-to-speech | 14.1M | $15 / 1M chars |
| 03 | Tidal 1.1 Pro | Bluefin Labs | image | 8.4M | $0.04 / image |
| 04 | Tidal Sketch 1 | Bluefin Labs | image | 6.2M | $0.02 / image |
| 05 | Maelstrom 3 | Tycho Systems | video | 1.2M | $0.35 / s |
Modalities
Five ways to route, one way to pay
Every modality hangs off the same base URL and the same key. Token models bill per token; the rest bill per unit of output.
Text
Chat, streaming, tool calls, structured output
Image
Generation across frontier diffusion models
Video
Text to video with length and resolution control
Speech to text
Transcription with timestamps and detection
Text to speech
Dozens of voices, natural pacing, 30+ languages
Methodology
How the numbers are computed
The leaderboard is generated from routed traffic, not submitted benchmarks. Here is exactly what each column means.
- Ranked by
- Routed tokens over a rolling 30 day window, counted after failover retries so a flaky provider does not double count.
- Latency figure
- Median time to first token across all providers serving the model, measured at the gateway edge.
- Refresh cadence
- The snapshot recomputes every Monday at 00:00 UTC and stays public for the full week.
- Price columns
- Provider list price per one million tokens, passed through with no markup.
Pick your models
Any row above is one model id away
Pass the id to the chat, image, video, or audio endpoint and the gateway handles provider selection, failover, and billing. Free tier covers every flagged row.