Model rankings: the most-used models by API calls
Every model on the API, ranked by the calls it answered in the last 7 days. Read at render from the public stats feed and joined to the live catalog for the developer and today's price.
Method: calls through the API in the last 7 days, read at render from GET https://api.relayswitch.io/v1/stats/top-models · generated 2026-11-02T07:40:20.187Z.
Models by calls
Rank, the model and its developer from the catalog, the calls counted by the feed, and the price it bills today in USD per million tokens.
| # | Model | Developer | Calls, 7 days | Input | Output |
|---|---|---|---|---|---|
| 1 | DeepSeek V4.1 Flash Abliterateddeepseek-v4.1-flash-abliterated | DeepSeek | 41,504 | $0.875 | $1.75 |
| 2 | GLM-5.3 Uncensoredglm-5.3-uncensored | Z.ai | 33,214 | $0.291 | $0.873 |
| 3 | GLM-5.3 Abliteratedglm-5.3-abliterated | Z.ai | 21,011 | $2.25 | $4.50 |
| 4 | GLM-5.3 Flash Uncensoredglm-5.3-flash-uncensored | Z.ai | 16,859 | $0.1086 | $0.3802 |
| 5 | GLM-5.3 Flash Abliteratedglm-5.3-flash-abliterated | Z.ai | 11,095 | $1.76 | $1.76 |
| 6 | DeepSeek V4.1 Flash 1017deepseek-v4.1-flash-1017 | DeepSeek | 9,013 | $0.25 | $1.20 |
| 7 | DeepSeek V4.1 Flash 1017 Uncensoreddeepseek-v4.1-flash-1017-uncensored | DeepSeek | 7,133 | $0.825 | $3.30 |
| 8 | DeepSeek V4 Pro Uncensoreddeepseek-v4-pro-uncensored | DeepSeek | 5,889 | $1.00 | $2.00 |
| 9 | GLM-5.3 Flash Apex Abliteratedglm-5.3-flash-apex-abliterated | Z.ai | 4,352 | $0.80 | $3.20 |
| 10 | DeepSeek V4.1 Flash Abliterated Fastdeepseek-v4.1-flash-abliterated-fast | DeepSeek | 4,308 | $1.75 | $3.50 |
| 11 | Kimi K3 Uncensoredkimi-k3-uncensored | Moonshot AI | 4,297 | $5.745 | $28.68 |
| 12 | Claude Opus 5.5claude-opus-5-5 | Anthropic | 3,971 | $2.80 | $10.00 |
| 13 | DeepSeek V4.1 Flash 1017 Uncensored Thinkingdeepseek-v4.1-flash-1017-uncensored-thinking | DeepSeek | 3,819 | $0.825 | $3.30 |
| 14 | GLM 5.3glm-5.3 | Z.ai | 3,388 | $0.3276 | $1.0296 |
| 15 | DeepSeek V4.1 Flash 0910 Uncensoreddeepseek-v4.1-flash-0910-uncensored | DeepSeek | 2,827 | $0.825 | $3.30 |
| 16 | Qwen3.8 27B Uncensoredqwen3.8-27b-uncensored | Qwen (community tune) | 2,536 | $1.25 | $3.75 |
| 17 | GPT-6.1 Solgpt-6.1-sol | OpenAI | 2,524 | $0.60 | $3.00 |
| 18 | GLM-5.3 Flash Code Uncensoredglm-5.3-flash-code-uncensored | Z.ai | 2,279 | $6.00 | $6.00 |
| 19 | Claude Haiku 5.5claude-haiku-5-5 | Anthropic | 2,217 | $0.06 | $0.15 |
| 20 | DeepSeek V4.1 Flashdeepseek-v4.1-flash | DeepSeek | 2,014 | $0.06 | $0.18 |
| 21 | GLM-5.3 Flash Cybersecurityglm-5.3-flash-cybersecurity | Z.ai | 1,973 | $0.525 | $1.75 |
| 22 | Gemini 4 Argongemini-4-argon | 1,969 | $0.60 | $3.00 | |
| 23 | Gemini 3.8 Flashgemini-3.8-flash | 1,855 | $0.4925 | $2.25 | |
| 24 | Claude Sonnet 5.5claude-sonnet-5-5 | Anthropic | 1,526 | $1.20 | $3.00 |
| 25 | MiniMax M3minimax-m3 | MiniMax | 1,361 | $0.09 | $0.36 |
| 26 | Kimi K3kimi-k3 | Moonshot AI | 1,325 | $0.90 | $4.50 |
| 27 | Nemotron 3 Ultra 550B Uncensorednemotron-3-ultra-550b-a55b-uncensored | NVIDIA | 1,111 | $1.125 | $6.60 |
| 28 | DeepSeek V4 Flash 0731deepseek-v4-flash-0731 | DeepSeek | 1,071 | $0.084 | $0.12 |
| 29 | GPT-6 Astragpt-6-astra | OpenAI | 1,033 | $3.00 | $15.00 |
| 30 | deepseek-v4.1-flash-redteam | — | 1,008 | — | — |
| 31 | GLM 5.3 Flashglm-5.3-flash | Z.ai | 980 | $0.09 | $0.15 |
| 32 | glm-5.3-redteam | — | 959 | — | — |
| 33 | Minimax M3.1 Flashminimax-m3.1-flash | MiniMax | 951 | $0.105 | $0.42 |
| 34 | Claude Haiku 4.5claude-haiku-4-5 | Anthropic | 899 | $0.60 | $3.00 |
| 35 | Claude Fable 5claude-fable-5 | Anthropic | 730 | $3.00 | $15.00 |
| 36 | Kimi K3.1 Prokimi-k3.1-pro | Moonshot AI | 704 | $1.00 | $5.00 |
| 37 | Claude Opus 5claude-opus-5 | Anthropic | 623 | $1.80 | $7.50 |
| 38 | MiMo V2.6 Flash Uncensoredmimo-v2.6-flash-uncensored | Xiaomi | 617 | $2.00 | $6.00 |
| 39 | Hermes 4 405Bhermes-4-405b | Nous Research | 608 | $1.20 | $4.80 |
| 40 | Gemini 2.5 Flashgemini-2.5-flash | 600 | $0.30 | $2.50 |
An id without a developer or price is one the feed counted that the catalog no longer lists. Prices are read from the catalog at render; the full list with cached-input rates is on the models page.
Any model on the list, on one key
Prepaid, per token, at the price in the table. Card or crypto, no KYC, no subscription.