AI Model Rankings
Real-world LLM usage on 3srouter. Models are ranked by tokens processed through the 3srouter API.
3srouter usage data · Through October 19, 2026. View source
Top models
Total tokens on 3srouter
LLM Leaderboard
| # | Model | Tokens | Change |
|---|---|---|---|
| 1 | DeepSeek V4.1 Flashdeepseek | 29.6T | -27% |
| 2 | 11.9T | -6% | |
| 3 | MiMo-V2.6-Flashxiaomi | 9.57T | -17% |
| 4 | Hy4 previewtencent | 7.82T | -6% |
| 5 | DeepSeek V4.1 Flash 1017deepseek | 7.34T | New |
| 6 | 5.29T | +33% | |
| 7 | DeepSeek V4 Flash 0731deepseek | 4.73T | +1% |
Top models by task
Tool calling
Models that connect language understanding with structured actions and external tools.
Programming
Find models for code generation, debugging, editing, and agentic development.
Creative writing
Explore models for storytelling, character interactions, and creative writing.
Vision
Analyze images, understand documents, and reason about visual content.
Fastest models
Observed throughput on 3srouter · October 20, 2026
| Model | Throughput | Latency | Providers |
|---|---|---|---|
| Qwen3.8 27B | 427 tok/s | 300 ms | 21 |
| GLM 5.2 | 402 tok/s | 327 ms | 34 |
| DeepSeek V4.1 Flash | 346 tok/s | 248 ms | 36 |
Cost per session
Compare the cost of real-world model sessions.
This breakdown is available in the live 3srouter rankings.
Open / Closed
Track the share of open-weight and proprietary model usage.
This breakdown is available in the live 3srouter rankings.
Languages
Explore model usage across natural languages.
This breakdown is available in the live 3srouter rankings.
Programming
Coding-capable models from the public catalog
MiniMax: MiniMax M3.1 Flash (free)minimax/minimax-m3.1-flash:free1M contextUnverified/M input tokens
Tencent: Hy4 previewtencent/hy4-preview1M contextUnverified/M input tokens
Qwen: Qwen3.8 Flashqwen/qwen3.8-flash1M context$0.15/M input tokensDomain
Explore how models are used across task domains.
This breakdown is available in the live 3srouter rankings.
Context Length
Xiaomi: MiMo-V2.6-Flashxiaomi/mimo-v2.6-flash1.1M contextUnverified/M input tokens
Xiaomi: MiMo-V2.6-Proxiaomi/mimo-v2.6-pro1.1M contextUnverified/M input tokens
Xiaomi: MiMo-V2.5-Proxiaomi/mimo-v2.5-pro1.1M contextUnverified/M input tokens
Xiaomi: MiMo-V2.5xiaomi/mimo-v2.51.1M contextUnverified/M input tokens
DeepSeek: DeepSeek V4.1 Flash 1017deepseek/deepseek-v4.1-flash-10171M context$0.15/M input tokensTool Calls
Explore models
MiniMax: MiniMax M3.1 Flash (free)minimax/minimax-m3.1-flash:free1M contextUnverified/M input tokens
DeepSeek: DeepSeek V4.1 Flash 1017deepseek/deepseek-v4.1-flash-10171M context$0.15/M input tokens
Qwen: Qwen3.8 Max Primeqwen/qwen3.8-max-prime1M contextUnverified/M input tokens
Xiaomi: MiMo-V2.6-Pro-UltraSpeedxiaomi/mimo-v2.6-pro-ultraspeed1M contextUnverified/M input tokensImages
Explore vision models
MiniMax: MiniMax M3.1 Flash (free)minimax/minimax-m3.1-flash:free1M contextUnverified/M input tokens
DeepSeek: DeepSeek V4.1 Flash 1017deepseek/deepseek-v4.1-flash-10171M context$0.15/M input tokens
Qwen: Qwen3.8 Max Primeqwen/qwen3.8-max-prime1M contextUnverified/M input tokens
Xiaomi: MiMo-V2.6-Pro-UltraSpeedxiaomi/mimo-v2.6-pro-ultraspeed1M contextUnverified/M input tokens
Xiaomi: MiMo-V2.6-Flashxiaomi/mimo-v2.6-flash1.1M contextUnverified/M input tokensTop Apps
View rankingsPublic rankings data: Original source · CC BY 4.0 · October 20, 2026.