Ranked by real token throughput on the OpenRouter marketplace - not benchmarks. Useful for seeing which models people actually run.
Tokens
114.75T
Requests
5.1B
Models
456
Lead
Hy4 preview
DeepSeek
Token mix this window
97% prompt · 3% completion
Share of token volume this window
12.8%
↑379%
11.3%
↑66%
10.8%
↑101%
10.8%
-
4.5%
-
4.4%
↑95%
3.5%
↓40%
3.2%
↓32%
2.6%
↑128%
2.0%
↓74%
Showing 20 of 456
| # | Model | Type | Tokens | Share | Requests | Tools | Price / M | Context | Δ wk |
|---|---|---|---|---|---|---|---|---|---|
| 1 | Hy4 preview Tencent · OpenRouter | Language | 14.7T 14.5T in · 140.6B out | 12.8% | 131.1M | - | $0.83 / $2.50 | 1.0M | ↑379% |
| 2 | GPT-5.6 Luna OpenAI · OpenRouter | Vision | 12.9T 12.7T in · 222.2B out | 11.3% | 418.8M | - | $0.20 / $1.20 | 1.1M | ↑66% |
| 3 | GLM 5.3 Flash Z.ai · OpenRouter | Vision | 12.4T 12.0T in · 420.4B out | 10.8% | 369.0M | - | $0.07 / $0.25 | 1.3M | ↑101% |
| 4 | DeepSeek V4 Flash 0731 DeepSeek · OpenRouter | Language | 12.4T 11.9T in · 416.8B out | 10.8% | 495.0M | - | $0.14 / $0.28 | 1.3M | - |
| 5 | DeepSeek V4 Flash 0423 DeepSeek · OpenRouter | Language | 5.19T 4.91T in · 273.8B out | 4.5% | 524.2M | - | $0.09 / $0.18 | 1.0M | - |
| 6 | MiniMax M3 (free) MiniMax · free · OpenRouter | Vision | 5.02T 4.97T in · 44.6B out | 4.4% | 66.5M | - | Free | 1.0M | ↑95% |
| 7 | Hy3 Tencent · OpenRouter | Language | 3.99T 3.93T in · 62.8B out | 3.5% | 56.4M | - | $0.13 / $0.53 | 262K | ↓40% |
| 8 | Nemotron 3 Ultra (free) NVIDIA · free · OpenRouter | Language | 3.64T 3.62T in · 19.5B out | 3.2% | 30.3M | - | Free | 1M | ↓32% |
| 9 | GLM 5.3 Z.ai · OpenRouter | Language | 3.02T 2.97T in · 55.1B out | 2.6% | 42.4M | - | $1.40 / $4.40 | 1.3M | ↑128% |
| 10 | MiMo-V2.5 Xiaomi · OpenRouter | Audio | 2.35T 2.34T in · 15.9B out | 2.0% | 70.1M | - | $0.14 / $0.28 | 1.1M | ↓74% |
| 11 | GLM 5.2 Z.ai · OpenRouter | Language | 2.35T 2.29T in · 59.4B out | 2.0% | 64.1M | - | $0.97 / $3.04 | 1.0M | ↓22% |
| 12 | Gemini 3.7 Flash Google · OpenRouter | Audio | 2.10T 2.05T in · 52.8B out | 1.8% | 65.4M | - | $0.75 / $3.75 | 1.0M | ↓47% |
| 13 | Kimi K3 Moonshot · OpenRouter | Vision | 1.98T 1.95T in · 30.2B out | 1.7% | 31.5M | - | $3.00 / $15 | 1.0M | ↑26% |
| 14 | GPT-5.6 Sol OpenAI · OpenRouter | Vision | 1.88T 1.86T in · 21.8B out | 1.6% | 32.4M | - | $2.00 / $10 | 1.1M | ↑11% |
| 15 | Claude Opus 5 Anthropic · OpenRouter | Vision | 1.78T 1.75T in · 30.8B out | 1.5% | 23.9M | - | $5.00 / $25 | 1M | ↑21% |
| 16 | MiniMax M3 MiniMax · OpenRouter | Vision | 1.45T 1.42T in · 24.2B out | 1.3% | 28.8M | - | $0.30 / $1.20 | 1.0M | ↓5% |
| 17 | Claude Sonnet 5 Anthropic · OpenRouter | Vision | 1.33T 1.31T in · 24.0B out | 1.2% | 26.2M | - | $2.00 / $10 | 1M | ↑11% |
| 18 | Laguna S 2.1 (free) Poolside · free · OpenRouter | Language | 1.33T 1.32T in · 10.3B out | 1.2% | 16.3M | - | Free | 262K | ↓2% |
| 19 | DeepSeek V4 Pro 0423 DeepSeek · OpenRouter | Language | 1.25T 1.21T in · 45.6B out | 1.1% | 52.9M | - | $1.04 / $2.07 | 1.0M | ↓29% |
| 20 | Solar Pro 4 Upstage · OpenRouter | Language | 1.21T 1.20T in · 14.5B out | 1.1% | 70.7M | - | $0.03 / $0.12 | 524K | ↑87% |
Weekly request share by model author
Models that actually invoke tools
Images processed through OpenRouter
Public apps by token volume this week
Hermes Agent
Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.
Claude Code
Claude Code is Anthropic's agentic coding tool that reads your entire codebase, plans and executes changes across files, runs tests, and iterates on failures, all from natural language prompts.
Kilo Code
Kilo Code is an open-source AI coding agent that works across VS Code, JetBrains, and CLI to help developers ship code faster with agentic workflows.
Cline
Cline is an open-source AI coding agent that lives inside your IDE, autonomously exploring your codebase, editing files, running terminal commands, and using browser automation.
Zazen (Freebuff fork)
pi
There are many coding agents, but this one is yours.
omp
OpenClaw
OpenClaw is an open-source AI agent that connects to your messaging apps and takes real actions on your behalf, from running commands and browsing the web to managing files and sending emails.
Codex
A coding agent that helps you build and ship with AI
DeepSeek Harness
Portkey AI
Control panel for AI apps
ISEKAI ZERO
AI adventures. Travel with your favorite characters
Source: OpenRouter rankings, as of 2026-09-06. Licensed under CC BY 4.0. Price is prompt / completion per million tokens. First-party app and direct vendor traffic are not included.