AI Gateway Models & Pricing

Every model each gateway advertises through its public model-list endpoint, unified into one matrix so you can see which gateways route a given model and what each charges per million tokens.

Pricing is not guaranteed to be accurate — rates change frequently, and this benchmark runs weekly, so treat this as a directional comparison and confirm current pricing with the gateway before making a decision.

Powered byNamespace· 4 vCPU / 16GB RAM· Northern Virginia, US
Last indexed: September 18
openrouter logoOpenrouter
442 models
49 authors
llmapi logoLlmapi
397 models
18 authors
vercel-ai-gateway logoVercel Ai Gateway
376 models
35 authors
pydantic-ai-gateway logoPydantic Ai Gateway
296 models
11 authors
llmgateway logoLlmgateway
270 models
24 authors
blazerail logoBlazerail
256 models
19 authors
concentrate-ai-gateway logoConcentrate Ai Gateway
187 models
20 authors
ngrok logoNgrok
137 models
11 authors
cloudflare-ai-gateway logoCloudflare Ai Gateway
130 models
1 author
novita logoNovita
120 models
25 authors
ramp logoRamp
70 models
3 authors
neon logoNeon
34 models
8 authors
We found 1,429 models
ModelCoverageContextLowest $/Mopenrouter logollmapi logovercel-ai-gateway logollmgateway logoblazerail logoconcentrate-ai-gateway logongrok logoramp logoneon logopydantic-ai-gateway logocloudflare-ai-gateway logonovita logo
GPT OSS 120B
OpenAI
11/12131K$0.032$0.14$0.15$0.6$0.15$0.6$0.1$0.5$0.032$0.14$0.15$0.6$0.037$0.17$0.15$0.6$0.15$0.6$0.15$0.6
GPT-5
OpenAI
11/12400K$1.25$10$1.25$10$1.25$10$1.25$10$1.25$10$1.25$10$1.25$10$1.25$10$1.25$10$1.25$10
GPT-5 Mini
OpenAI
11/12400K$0.25$2$0.25$2$0.25$2$0.25$2$0.25$2$0.25$2$0.25$2$0.25$2$0.25$2$0.25$2
GPT-5.2
OpenAI
11/12400K$1.75$14$1.75$14$1.75$14$1.75$14$1.75$14$1.75$14$1.75$14$1.75$14$1.75$14$1.75$14
GPT-5.4
OpenAI
11/121.1M$2.5$15$2.5$15$2.5$15$2.5$15$2.5$15$2.5$15$2.5$15$2.5$15$2.5$15$2.5$15
GPT-5.4 Mini
OpenAI
11/12400K$0.75$4.5$0.75$4.5$0.75$4.5$0.75$4.5$0.75$4.5$0.75$4.5$0.75$4.5$0.75$4.5$0.75$4.5$0.75$4.5
GPT-5.6 Luna
OpenAI
11/121.1M$0.06$0.36$0.2$1.2$0.2$1.2$0.2$1.2$0.2$1.2$0.06$0.36$0.2$1.2$0.2$1.2$0.2$1.2$0.2$1.2
GPT-5.6 Sol
OpenAI
11/121.1M$1.75$10$2$10$5$30$2$10$4$20$1.75$10.5$4$20$4$20$4$20$5$30
GPT-5.6 Terra
OpenAI
11/121.1M$0.7$4.2$2$12$2$12$2$12$2$12$0.7$4.2$2$12$2$12$2$12$2$12
GPT-6 Astra
OpenAI
11/121.1M$7$35$10$50$10$50$10$50$10$50$7$35$10$50$10$50$10$50$10$50
Kimi K3
Moonshot
11/121M$2.1$10.95$2.1$10.95$3$15$3$15$3$15$3$15$2.83$14.13$3$15$3$15$3$15
DeepSeek V4 Pro
DeepSeek
10/121.1M$0.435$0.87$1.6$3.2$1.74$3.48$0.66$1.98$0.435$0.87$1.3$2.6$1.3$2.6$0.66$1.98$1.74$3.48$1.74$3.48
GPT-4.1
OpenAI
10/121M$2$8$2$8$2$8$2$8$2$8$2$8$2$8$2$8$2$8
GPT-4.1 Mini
OpenAI
10/121M$0.4$1.6$0.4$1.6$0.4$1.6$0.4$1.6$0.4$1.6$0.4$1.6$0.4$1.6$0.4$1.6$0.4$1.6
GPT-4o
OpenAI
10/12128K$2.5$10$2.5$10$2.5$10$2.5$10$2.5$10$2.5$10$2.5$10$2.5$10$2.5$10
GPT-4o Mini
OpenAI
10/12128K$0.15$0.6$0.15$0.6$0.15$0.6$0.15$0.6$0.15$0.6$0.15$0.6$0.15$0.6$0.15$0.6$0.15$0.6
GPT-5 Nano
OpenAI
10/12400K$0.05$0.4$0.05$0.4$0.05$0.4$0.05$0.4$0.05$0.4$0.05$0.4$0.05$0.4$0.05$0.4$0.05$0.4
GPT-5.1
OpenAI
10/12400K$1.25$10$1.25$10$1.25$10$1.25$10$1.25$10$1.25$10$1.25$10$1.25$10$1.25$10
GPT-5.3-Codex
OpenAI
10/12400K$1.75$14$1.75$14$1.75$14$1.75$14$1.75$14$1.75$14$1.75$14$1.75$14$1.75$14
GPT-5.4 Nano
OpenAI
10/12400K$0.2$1.25$0.2$1.25$0.2$1.25$0.2$1.25$0.2$1.25$0.2$1.25$0.2$1.25$0.2$1.25$0.2$1.25
GPT-5.5
OpenAI
10/121.1M$5$30$5$30$5$30$5$30$5$30$5$30$5$30$5$30$5$30
Claude Fable 5
Anthropic
9/121M$10$50$10$50$10$50$10$50$10$50$10$50$10$50$10$50$10$50
Claude Fable 5.1
Anthropic
9/121M$10$50$10$50$10$50$10$50$10$50$10$50$10$50$10$50$10$50
Claude Haiku 4.5
Anthropic
9/12200K$1$5$1$5$1$5$1$5$1$5$1$5$1$5$1$5$1$5
Claude Opus 4.6
Anthropic
9/121M$5$25$5$25$5$25$5$25$5$25$5$25$5$25$5$25$5$25
Claude Opus 4.8
Anthropic
9/121M$5$25$5$25$5$25$5$25$5$25$5$25$5$25$5$25$5$25
Claude Opus 5
Anthropic
9/121M$5$25$5$25$5$25$5$25$5$25$5$25$5$25$5$25$5$25
Claude Sonnet 4.5
Anthropic
9/121M$3$15$3$15$3$15$3$15$3$15$3$15$3$15$3$15$3$15
Claude Sonnet 4.6
Anthropic
9/121M$3$15$3$15$3$15$3$15$3$15$3$15$3$15$3$15$3$15
DeepSeek V4 Flash
DeepSeek
9/121.1M$0.0498$0.0997$0.0498$0.0997$0.19$0.51$0.13$0.26$0.05$0.1$0.09$0.18$0.22$0.66$0.14$0.28$0.14$0.28
DeepSeek V4 Flash 0731
DeepSeek
9/121.3M$0.06$0.12$0.06$0.12$0.14$0.28$0.076$0.153$0.08$0.18$0.22$0.66$0.22$0.66$0.22$0.66
Gemini 3.5 Flash
Google
9/121M$1.5$9$1.5$9$1.5$9$1.5$9$1.5$9$1.5$9$1.5$9$1.5$9$1.5$9
Gemini 3.6 Flash
Google
9/121M$0.75$3.75$0.75$3.75$1.5$7.5$0.75$3.75$0.75$3.75$0.75$3.75$0.75$3.75$0.75$3.75$1.5$7.5
GLM-5.3-Flash
Z.ai
9/121.3M$0.075$0.25$0.09$0.3$0.15$0.5$0.15$0.5$0.088$0.25$0.15$0.5$0.075$0.25$0.15$0.5$0.15$0.5
GPT OSS 20B
OpenAI
9/12131K$0.03$0.13$0.03$0.13$0.07$0.3$0.03$0.14$0.04$0.19$0.03$0.14$0.075$0.3$0.07$0.3
GPT-3.5 Turbo
OpenAI
9/1216K$0.5$1.5$0.5$1.5$0.5$1.5$0.5$1.5$0.5$1.5$0.5$1.5$0.5$1.5$0.5$1.5
GPT-4 Turbo
OpenAI
9/12128K$10$30$10$30$10$30$10$30$10$30$10$30$10$30$10$30
GPT-4.1 Nano
OpenAI
9/121M$0.1$0.4$0.1$0.4$0.1$0.4$0.1$0.4$0.1$0.4$0.1$0.4$0.1$0.4$0.1$0.4
GPT-5.2 Pro
OpenAI
9/12400K$21$168$21$168$21$168$21$168$21$168$21$168$21$168$21$168
GPT-5.4 Pro
OpenAI
9/121.1M$30$180$30$180$30$180$30$180$30$180$30$180$30$180$30$180
Kimi K2.6
Moonshot
9/12262K$0.6$3.05$0.95$4$0.95$4$0.95$4$0.6$3.05$0.95$4$0.75$3.5$0.95$4$0.858$3.57
MiniMax M3
MiniMax
9/121M$0.28$1.1$0.3$1.2$0.6$2.4$0.3$1.2$0.3$1.2$0.28$1.1$0.28$1.1$0.3$1.2$0.3$1.2
o1
OpenAI
9/12200K$15$60$15$60$15$60$15$60$15$60$15$60$15$60$15$60
o3
OpenAI
9/12200K$2$8$2$8$2$8$2$8$2$8$2$8$2$8$2$8
o3 Mini
OpenAI
9/12200K$1.1$4.4$1.1$4.4$1.1$4.4$1.1$4.4$1.1$4.4$1.1$4.4$1.1$4.4$1.1$4.4
Claude Opus 4.7
Anthropic
8/121M$5$25$5$25$5$25$5$25$5$25$5$25$5$25$5$25
Claude Sonnet 5
Anthropic
8/121M$2$10$2$10$2$10$2$10$2$10$2$10$2$10$2$10
DeepSeek V4.1 Flash
DeepSeek
8/121.1M$0.15$0.6$0.15$0.6$0.3$1.2$0.3$1.2$0.15$0.6$0.2$0.6$0.15$0.6$0.22$0.66
Gemini 2.5 Flash
Google
8/121M$0.3$2.5$0.3$2.5$0.3$2.5$0.3$2.5$0.3$2.5$0.3$2.5$0.3$2.5$0.3$2.5
Gemini 2.5 Flash Lite
Google
8/121M$0.1$0.4$0.1$0.4$0.1$0.4$0.1$0.4$0.1$0.4$0.1$0.4$0.1$0.4$0.1$0.4
Gemini 2.5 Pro
Google
8/121M$1.25$10$1.25$10$1.25$10$1.25$10$1.25$10$1.25$10$1.25$10$1.25$10
Gemini 3.1 Pro Preview
Google
8/121M$2$12$2$12$2$12$2$12$2$12$2$12$2$12$2$12
Gemini 3.5 Flash Lite
Google
8/121M$0.3$2.5$0.3$2.5$0.3$2.5$0.3$2.5$0.3$2.5$0.3$2.5$0.3$2.5$0.3$2.5
Gemini 3.7 Flash
Google
8/121M$0.75$3.75$0.75$3.75$0.75$3.75$0.75$3.75$0.75$3.75$0.75$3.75$0.75$3.75$0.75$3.75
Gemini 3.8 Flash
Google
8/121M$0.75$3.75$0.75$3.75$0.75$3.75$0.75$3.75$0.75$3.75$0.75$3.75$0.75$3.75$0.75$3.75
GLM-5.1
Z.ai
8/12205K$0.931$2.93$0.966$3.04$1.4$4.4$1.4$4.4$0.931$2.93$1.23$3.86$1.4$4.4$1.4$4.4
GLM-5.2
Z.ai
8/121M$0.44$1.4$0.5544$1.74$0.8$2.55$0.8$2.55$0.44$1.4$1.4$4.4$1.4$4.4$1.4$4.4
GLM-5.3
Z.ai
8/121.3M$0.77$2.42$1.4$4.4$1.4$4.4$1.4$4.4$1.2$4$0.77$2.42$1.4$4.4$1.4$4.4
GPT-4
OpenAI
8/128K$30$60$30$60$30$60$30$60$30$60$30$60$30$60
GPT-5 Pro
OpenAI
8/12400K$15$120$15$120$15$120$15$120$15$120$15$120$15$120

Want to see a gateway added?

Let us know on X

Methodology

This page is a catalog snapshot, not a latency benchmark. It reflects what each gateway's model-list endpoint returned at index time, and prices are whatever that endpoint reported — a gateway with no pricing in its catalog is shown as listed-but-unpriced, not free. Pricing is not guaranteed to be accurate: gateways can change rates faster than the next index run, so figures here may lag or diverge from what a gateway actually charges — always verify current pricing directly with the gateway before relying on it. For gateway latency and throughput see the AI Gateway Benchmarks.

What We Collect

One authenticated GET against each gateway's public model-list endpoint (/v1/models or its equivalent), recording the HTTP status, response time, and every model entry returned. Gateways that don't expose a model list are still attempted so the matrix is honest about which ones support programmatic discovery.

From each entry we keep the ID, display name, owner, context window, max output tokens, and pricing. Per-token prices are normalized to USD per 1M tokens. Entries priced per second, per image, per request, or per character (video, speech, and image models) show that unit's rate directly rather than being forced into a token price.

How Models Are Matched

Gateways spell the same model differently: anthropic/claude-fable-5.1, claude-fable-5-1, gpt-5.4-mini-2026-03-17. Each ID is reduced to a canonical key by dropping the creator prefix, unifying ., :, and _ to -, and stripping a trailing date stamp. Other suffixes (:batch, -thinking) are kept, so distinct SKUs stay as separate rows.

A row's author is the model creator most gateways agree on, ignoring serving providers like azure or system. Hover any price to see the exact IDs that gateway uses and how far its price sits above the lowest listed. Matching is heuristic — an unexpected merge or split is a bug worth reporting.

PartnersLatitude