Ranked by real token throughput on the OpenRouter marketplace - not benchmarks. Useful for seeing which models people actually run.
Tokens
166.70T
Requests
7.6B
Models
517
Lead
DeepSeek V4.1 Flash
DeepSeek
Token mix this window
97% prompt · 3% completion
Share of token volume this window
24.6%
↑60%
7.3%
↑26%
6.9%
↑19%
5.3%
-
5.0%
↑27%
4.2%
↑15%
4.1%
↑22%
3.0%
↑104%
2.9%
↓18%
2.3%
↑29%
Showing 20 of 517
| # | Model | Type | Tokens | Share | Requests | Tools | Price / M | Context | Δ wk |
|---|---|---|---|---|---|---|---|---|---|
| 1 | DeepSeek V4.1 Flash DeepSeek · OpenRouter | Vision | 41.0T 40.2T in · 780.8B out | 24.6% | 705.2M | - | $0.15 / $0.60 | 1.0M | ↑60% |
| 2 | GLM 5.3 Flash Z.ai · OpenRouter | Vision | 12.2T 11.8T in · 467.7B out | 7.3% | 722.8M | - | $0.02 / $1.20 | 1.0M | ↑26% |
| 3 | MiMo-V2.6-Flash Xiaomi · OpenRouter | Audio | 11.5T 11.3T in · 136.3B out | 6.9% | 142.4M | - | $0.14 / $0.28 | 1.1M | ↑19% |
| 4 | Step 5 Preview StepFun · OpenRouter | Vision | 8.84T 8.78T in · 51.9B out | 5.3% | 47.7M | - | $1.00 / $2.70 | 1M | - |
| 5 | Hy4 preview Tencent · OpenRouter | Language | 8.28T 8.20T in · 77.3B out | 5.0% | 74.8M | - | $0.75 / $2.25 | 1.0M | ↑27% |
| 6 | GPT-6 Luna OpenAI · OpenRouter | Vision | 7.08T 6.87T in · 210.7B out | 4.2% | 344.0M | - | $0.10 / $0.50 | 1.1M | ↑15% |
| 7 | Nemotron 3 Ultra (free) NVIDIA · free · OpenRouter | Language | 6.87T 6.84T in · 32.0B out | 4.1% | 51.4M | - | Free | 1M | ↑22% |
| 8 | Claude Opus 5.5 Anthropic · OpenRouter | Vision | 5.02T 4.97T in · 53.6B out | 3.0% | 43.9M | - | $4.00 / $20 | 1M | ↑104% |
| 9 | DeepSeek V4 Flash 0731 DeepSeek · OpenRouter | Language | 4.92T 4.71T in · 202.8B out | 2.9% | 373.7M | - | $0.02 / $0.64 | 1.0M | ↓18% |
| 10 | GLM 5.3 Z.ai · OpenRouter | Language | 3.79T 3.73T in · 59.0B out | 2.3% | 71.6M | - | $0.08 / $4.80 | 1.0M | ↑29% |
| 11 | jev-1.13-20260917 Typesafe · OpenRouter | Language | 3.36T 3.07T in · 291.1B out | 2.0% | 1.0B | - | Free | - | ↑8% |
| 12 | DeepSeek V4 Flash 0423 DeepSeek · OpenRouter | Language | 3.07T 2.88T in · 193.3B out | 1.8% | 334.3M | - | $0.01 / $1.28 | 1.0M | ↓5% |
| 13 | Hy3 Tencent · OpenRouter | Language | 2.51T 2.47T in · 41.1B out | 1.5% | 33.2M | - | $0.08 / $0.33 | 262K | ↑21% |
| 14 | Claude Sonnet 5.5 Anthropic · OpenRouter | Vision | 2.43T 2.40T in · 30.8B out | 1.5% | 31.4M | - | $2.00 / $10 | 1M | ↑208% |
| 15 | Vision | 2.27T 2.23T in · 38.5B out | 1.4% | 42.7M | - | $0.10 / $0.20 | 1.0M | ↑58% | |
| 16 | Kimi K3 Moonshot · OpenRouter | Vision | 2.15T 2.12T in · 33.4B out | 1.3% | 33.3M | - | $0.73 / $10 | 1.0M | ↑33% |
| 17 | GPT-6.1 Sol OpenAI · OpenRouter | Vision | 2.12T 2.09T in · 32.7B out | 1.3% | 46.1M | - | $2.00 / $10 | 1.1M | ↑191% |
| 18 | Gemini 3.8 Flash Google · OpenRouter | Audio | 2.05T 1.97T in · 71.3B out | 1.2% | 81.5M | - | $0.75 / $3.75 | 1.0M | ↓6% |
| 19 | GPT-5.6 Luna OpenAI · OpenRouter | Vision | 1.84T 1.77T in · 65.6B out | 1.1% | 158.9M | - | $0.20 / $1.20 | 1.1M | ↓58% |
| 20 | GPT-5.6 Sol OpenAI · OpenRouter | Vision | 1.57T 1.54T in · 26.1B out | 0.9% | 32.3M | - | $2.00 / $10 | 1.1M | ↑18% |
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.
Codex
A coding agent that helps you build and ship with AI
Freebuff
Powerful coding models, funded by ads
pi
There are many coding agents, but this one is yours.
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.
omp
DeepSeek Harness
OpenHands
AI coding agent that can run commands, browse the web, call APIs
StepCode
Source: OpenRouter rankings, as of 2026-10-11. Licensed under CC BY 4.0. Price is prompt / completion per million tokens. First-party app and direct vendor traffic are not included.