Skip to content

Which gateways overlap, and which are cheaper

How much of each gateway's catalog you would still have on another one, which models only one gateway carries, and — where both publish a price for the same model — which is cheaper more often.

Shared catalog

Each cell is the number of models both gateways carry. The diagonal is the gateway's own catalog size, not an overlap.

GatewayOpenRouterVercelCloudflareOrcaRouterRequestyConcentrateLLM GatewayBasetenDarkbloomRampTogetherSailFireworksMorphCheaper InferenceSayGMExperiential LabsNanoGPTIsoquantceleris
OpenRouter 4521694461171445548105301981196438125100
Vercel AI Gateway 169498237105135455114233187121015524215700
Cloudflare AI Gateway 4462371,982153203686910534298131010038829300
OrcaRouter 1171051532149533358227148873933010800
Requesty 144135203956293633113321471363723713600
Concentrate 554568333620811715685130145251245420
LLM Gateway 4851693533117288046610130188251304520
Baseten 10141081110200786861011100
Darkbloom 525235401201000424510
Ramp Router 303334273268667010953647618723520
Together AI 19182914145108158258511061600
Sail 878871160351045111800
Fireworks AI 11121381333806842066041100
Morph 91010760060455614000700
Cheaper Inference 641551003937145188147611160570271696830
SayGM 32332252502180100274525320
Experiential Labs 814288303712413014726140169252756720
NanoGPT 2511572931081365445115351681176836762500
Isoquant 00000220120000322030
Celeris 00000000000000000002

Two gateways can carry the same model under different ids — Concentrate and LLM Gateway publish bare ids, which only align to the canonical form when the match is unambiguous. Where alignment is uncertain the model is counted separately, so these numbers are a floor rather than a ceiling.

Price head-to-head

Counted only where both gateways carry the model and both publish an input price. Ties are common, because gateways often pass a provider's price through unchanged.

Pair Shared Both priced First cheaper Second cheaper Same price Verdict
OpenRouter vs NanoGPT 251 251 34 53 164 NanoGPT cheaper more often
LLM Gateway vs Cheaper Inference 188 188 24 149 15 Cheaper Inference cheaper more often
Cheaper Inference vs Experiential Labs 169 166 148 18 0 Cheaper Inference cheaper more often
OpenRouter vs Vercel AI Gateway 169 169 32 13 124 OpenRouter cheaper more often
Vercel AI Gateway vs NanoGPT 157 157 13 43 101 NanoGPT cheaper more often
Vercel AI Gateway vs Cheaper Inference 155 112 11 99 2 Cheaper Inference cheaper more often
OpenRouter vs Requesty 144 144 37 15 92 OpenRouter cheaper more often
Requesty vs NanoGPT 136 136 8 48 80 NanoGPT cheaper more often
Vercel AI Gateway vs Requesty 135 135 19 15 101 Vercel AI Gateway cheaper more often
LLM Gateway vs Experiential Labs 130 127 28 17 82 LLM Gateway cheaper more often
OpenRouter vs OrcaRouter 117 115 25 13 77 OpenRouter cheaper more often
OrcaRouter vs NanoGPT 108 108 12 30 66 NanoGPT cheaper more often
Vercel AI Gateway vs OrcaRouter 105 102 9 6 87 Vercel AI Gateway cheaper more often
OrcaRouter vs Requesty 95 95 13 8 74 OrcaRouter cheaper more often
OpenRouter vs Experiential Labs 81 80 16 3 61 OpenRouter cheaper more often
Ramp Router vs Cheaper Inference 76 76 1 75 0 Cheaper Inference cheaper more often
Ramp Router vs Experiential Labs 72 72 4 9 59 Experiential Labs cheaper more often
Cheaper Inference vs NanoGPT 68 68 51 16 1 Cheaper Inference cheaper more often
Experiential Labs vs NanoGPT 67 64 10 16 38 NanoGPT cheaper more often
LLM Gateway vs Ramp Router 66 66 12 0 54 LLM Gateway cheaper more often
OpenRouter vs Cheaper Inference 64 64 13 50 1 Cheaper Inference cheaper more often
Vercel AI Gateway vs LLM Gateway 51 43 2 10 31 LLM Gateway cheaper more often
OpenRouter vs LLM Gateway 48 47 11 10 26 OpenRouter cheaper more often
LLM Gateway vs NanoGPT 45 45 9 9 27 Evenly split
Vercel AI Gateway vs Experiential Labs 42 40 4 1 35 Vercel AI Gateway cheaper more often
OrcaRouter vs Cheaper Inference 39 39 6 33 0 Cheaper Inference cheaper more often
Requesty vs Cheaper Inference 37 37 2 35 0 Cheaper Inference cheaper more often
Requesty vs Experiential Labs 37 36 5 5 26 Evenly split
OrcaRouter vs LLM Gateway 35 35 6 8 21 LLM Gateway cheaper more often
Ramp Router vs NanoGPT 35 35 0 18 17 NanoGPT cheaper more often
Requesty vs LLM Gateway 33 33 7 5 21 Requesty cheaper more often
Vercel AI Gateway vs Ramp Router 33 33 14 1 18 Vercel AI Gateway cheaper more often
Requesty vs Ramp Router 32 32 13 0 19 Requesty cheaper more often
OpenRouter vs Ramp Router 30 30 17 0 13 OpenRouter cheaper more often
OrcaRouter vs Experiential Labs 30 30 7 3 20 OrcaRouter cheaper more often
Cheaper Inference vs SayGM 27 27 12 15 0 SayGM cheaper more often
OrcaRouter vs Ramp Router 27 27 16 0 11 OrcaRouter cheaper more often
LLM Gateway vs SayGM 25 25 5 20 0 SayGM cheaper more often
SayGM vs Experiential Labs 25 25 25 0 0 SayGM cheaper more often
OpenRouter vs Together AI 19 19 10 1 8 OpenRouter cheaper more often
Ramp Router vs SayGM 18 18 0 18 0 SayGM cheaper more often
Vercel AI Gateway vs Together AI 18 14 3 3 8 Evenly split
Together AI vs NanoGPT 16 16 1 12 3 NanoGPT cheaper more often
OrcaRouter vs Together AI 14 14 6 3 5 OrcaRouter cheaper more often
Requesty vs Together AI 14 14 2 5 7 Together AI cheaper more often
Vercel AI Gateway vs Baseten 14 14 3 1 10 Vercel AI Gateway cheaper more often
Baseten vs NanoGPT 11 11 1 9 1 NanoGPT cheaper more often
Requesty vs Baseten 11 11 2 0 9 Requesty cheaper more often
Together AI vs Cheaper Inference 11 11 0 7 4 Cheaper Inference cheaper more often
LLM Gateway vs Together AI 10 10 3 1 6 LLM Gateway cheaper more often
OpenRouter vs Baseten 10 10 7 1 2 OpenRouter cheaper more often
Vercel AI Gateway vs Morph 10 9 2 5 2 Morph cheaper more often
OpenRouter vs Morph 9 8 4 2 2 OpenRouter cheaper more often
Baseten vs Together AI 8 8 1 0 7 Baseten cheaper more often
OrcaRouter vs Baseten 8 8 4 1 3 OrcaRouter cheaper more often
Baseten vs Ramp Router 7 7 0 0 7 Evenly split
Morph vs NanoGPT 7 6 4 2 0 Morph cheaper more often
OrcaRouter vs Morph 7 6 0 6 0 Morph cheaper more often
Baseten vs Morph 6 6 1 5 0 Morph cheaper more often
Requesty vs Morph 6 6 1 5 0 Morph cheaper more often
Together AI vs Experiential Labs 6 6 1 2 3 Experiential Labs cheaper more often
Darkbloom vs NanoGPT 5 5 2 2 1 Evenly split
OpenRouter vs Darkbloom 5 5 0 4 1 Darkbloom cheaper more often
Ramp Router vs Together AI 5 5 0 0 5 Evenly split
Together AI vs Morph 5 5 0 5 0 Morph cheaper more often
Darkbloom vs Cheaper Inference 4 4 3 1 0 Darkbloom cheaper more often
Darkbloom vs Experiential Labs 4 4 3 1 0 Darkbloom cheaper more often
LLM Gateway vs Darkbloom 4 4 0 4 0 Darkbloom cheaper more often
Ramp Router vs Morph 4 4 0 4 0 Morph cheaper more often
Cheaper Inference vs Isoquant 3 3 1 2 0 Isoquant cheaper more often
OpenRouter vs SayGM 3 3 0 3 0 SayGM cheaper more often
OrcaRouter vs SayGM 3 3 0 3 0 SayGM cheaper more often
Requesty vs Darkbloom 3 3 0 3 0 Darkbloom cheaper more often
SayGM vs NanoGPT 3 3 3 0 0 SayGM cheaper more often
Darkbloom vs SayGM 2 2 2 0 0 Darkbloom cheaper more often
Experiential Labs vs Isoquant 2 2 0 2 0 Isoquant cheaper more often
LLM Gateway vs Isoquant 2 2 0 2 0 Isoquant cheaper more often
OrcaRouter vs Darkbloom 2 2 1 1 0 Evenly split
Ramp Router vs Isoquant 2 2 0 2 0 Isoquant cheaper more often
Requesty vs SayGM 2 2 0 2 0 SayGM cheaper more often
SayGM vs Isoquant 2 2 0 2 0 Isoquant cheaper more often
Vercel AI Gateway vs Darkbloom 2 2 0 2 0 Darkbloom cheaper more often
Vercel AI Gateway vs SayGM 2 2 0 2 0 SayGM cheaper more often
Baseten vs Cheaper Inference 1 1 0 1 0 Cheaper Inference cheaper more often
Baseten vs Experiential Labs 1 1 0 0 1 Evenly split
Darkbloom vs Together AI 1 1 1 0 0 Darkbloom cheaper more often
Darkbloom vs Isoquant 1 1 1 0 0 Darkbloom cheaper more often

Pairs that cannot be compared on price

These pairs share models, but at least one of them publishes no price for any of them — so there is nothing to compare. An absent price is not a cheap one.

OpenRouter + Cloudflare 446 Cloudflare + NanoGPT 293 Vercel + Cloudflare 237 Cloudflare + Requesty 203 Cloudflare + OrcaRouter 153 Concentrate + Cheaper Inference 145 Concentrate + Experiential Labs 124 Concentrate + LLM Gateway 117 Cloudflare + Cheaper Inference 100 Cloudflare + Experiential Labs 88 Cloudflare + LLM Gateway 69 Cloudflare + Concentrate 68 Concentrate + Ramp 68 OpenRouter + Concentrate 55 Concentrate + NanoGPT 54 Vercel + Concentrate 45 Requesty + Concentrate 36 Cloudflare + Ramp 34 OrcaRouter + Concentrate 33 Cloudflare + Together 29 Concentrate + SayGM 25 Cloudflare + Fireworks 13 Requesty + Fireworks 13 Vercel + Fireworks 12 Fireworks + NanoGPT 11 OpenRouter + Fireworks 11 Cloudflare + Baseten 10 Cloudflare + Morph 10 Baseten + Fireworks 8 Cloudflare + Sail 8 OpenRouter + Sail 8 OrcaRouter + Sail 8 OrcaRouter + Fireworks 8 Sail + NanoGPT 8 Together + Fireworks 8 Requesty + Sail 7 Vercel + Sail 7 Baseten + Sail 6 Fireworks + Morph 6 Fireworks + Cheaper Inference 6 Ramp + Fireworks 6 Cloudflare + Darkbloom 5 Concentrate + Darkbloom 5 Concentrate + Together 5 Sail + Morph 5 Together + Sail 5 Fireworks + Experiential Labs 4 Sail + Fireworks 4 Cloudflare + SayGM 3 Concentrate + Fireworks 3 LLM Gateway + Fireworks 3 Ramp + Sail 3 Concentrate + Isoquant 2 Concentrate + Baseten 1 Concentrate + Sail 1 LLM Gateway + Sail 1 Sail + Cheaper Inference 1 Sail + SayGM 1 Sail + Experiential Labs 1

Models only one gateway carries

The part of a gateway's catalog you would lose entirely by leaving it. Those exclusive models are very unevenly distributed across gateways.

Cloudflare 1,342 Requesty 412 NanoGPT 316 Cheaper Inference 211 Vercel 161 LLM Gateway 75 Experiential Labs 55 Together 52 OrcaRouter 51 Concentrate 29 SayGM 18 Ramp 13 OpenRouter 6 Baseten 5 Darkbloom 5 Morph 3 Sail 2 Fireworks 2 celeris 2 Isoquant 0

Cloudflare's figure is dominated by its Workers AI catalog — largely image, video, and embedding models that no other tracked gateway resells.