Themed ranking
Best for agentic workloads
Ranked by real agentic-task usage share, weighted across five real OpenRouter agent tags (tool dispatch, workflow execution, planning, and more).
As of August 9, 2026, DeepSeek: DeepSeek V4 Flash 0423 ranks #1 for best for agentic workloads at agentic traffic share 38.3%.
- #1 DeepSeek: DeepSeek V4 Flash 0423 - Agentic traffic share 38.3%
- #2 Tencent: Hy3 - Agentic traffic share 14.5%
- #3 OpenAI: GPT-5.6 Luna (batch) - Agentic traffic share 8.8%
- #4 DeepSeek: DeepSeek V4 Pro - Agentic traffic share 7.7%
- #5 Xiaomi: MiMo-V2.5 - Agentic traffic share 7.6%
- #6 Z.ai: GLM 5.2 (batch) - Agentic traffic share 6.8%
- #7 Google: Gemini 2.5 Flash Lite (batch) - Agentic traffic share 4.3%
- #8 Google: Gemini 3.6 Flash (batch) - Agentic traffic share 3.0%
- #9 NVIDIA: Nemotron 3 Ultra (free) - Agentic traffic share 2.7%
- #10 DeepSeek: DeepSeek V3.1 Terminus - Agentic traffic share 1.8%
| # | Model | Agentic traffic share |
|---|---|---|
| 01 | DeepSeek: DeepSeek V4 Flash 0423 | 38.3% |
| 02 | Tencent: Hy3 | 14.5% |
| 03 | OpenAI: GPT-5.6 Luna (batch) | 8.8% |
| 04 | DeepSeek: DeepSeek V4 Pro | 7.7% |
| 05 | Xiaomi: MiMo-V2.5 | 7.6% |
| 06 | Z.ai: GLM 5.2 (batch) | 6.8% |
| 07 | Google: Gemini 2.5 Flash Lite (batch) | 4.3% |
| 08 | Google: Gemini 3.6 Flash (batch) | 3.0% |
| 09 | NVIDIA: Nemotron 3 Ultra (free) | 2.7% |
| 10 | DeepSeek: DeepSeek V3.1 Terminus | 1.8% |
| 11 | OpenAI: gpt-oss-20b (free) | 1.2% |
| 12 | MoonshotAI: Kimi K3 | 0.9% |
| 13 | Google: Gemini 3 Flash Preview (batch) | 0.8% |
| 14 | Poolside: Laguna S 2.1 (free) | 0.5% |
| 15 | Google: Gemini 3.5 Flash Lite (batch) | 0.4% |
| 16 | OpenAI: gpt-oss-120b | 0.3% |
| 17 | Perplexity: Sonar | 0.2% |
| 18 | MiniMax: MiniMax M3 (batch) | 0.2% |
Source: OpenRouter (openrouter.ai/rankings), as of August 9, 2026.
Benchmark scores: Artificial Analysis (artificialanalysis.ai) via OpenRouter (openrouter.ai/rankings).
Methodology: smophy.ai/benchmark/methodology
Token counts originate from each provider's own tokenizer and are not directly comparable across providers.
