Agenttaken-ranglijst
Zie welke AI-modellen het best presteren op Agenttaken, welke betrouwbaar blijven en waar de grootste verschillen zitten. Sorteren op: Totale kosten ↑.
514/514
Modellen filteren
Geen modellen komen overeen met de huidige zoekopdracht en filters.
| Rang | Model | Bedrijf | Agenttaken-score | Score | Totale kosten | Correcte tests | Responstijd (gem.) |
|---|---|---|---|---|---|---|---|
| #198 | Inkling Small high | Thinkingmachines | 0.3 | 0.7 | $0.398 | 0/1 | 23ms |
| #134 | Qwen3.7 Plus medium | Qwen | 1.0 | 0.8 | $0.415 | 1/1 | 196.5s |
| #258 | MiniMax M3 medium | Minimax | 0.3 | 0.7 | $0.422 | 0/1 | 116.5s |
| #290 | Beam 1.0 501B A23B low | Reflection | 0.3 | 0.6 | $0.432 | 0/1 | 29.5s |
| #180 | Qwen3.5 Plus 2026-04-20 medium | Qwen | 1.0 | 0.8 | $0.435 | 1/1 | 203.0s |
| #154 | GPT-5.6 Terra low | OpenAI | 1.0 | 0.8 | $0.435 | 1/1 | 54.6s |
| #178 | Nemotron 3.5 Super VL medium | NVIDIA | 1.0 | 0.8 | $0.437 | 1/1 | 113.6s |
| #140 | Gemini 4 Argon low | 1.0 | 0.8 | $0.438 | 1/1 | 60.9s | |
| #445 | MiniMax M2.5 medium | Minimax | 0.3 | 0.4 | $0.440 | 0/1 | 600.8s |
| #208 | Gemini 3.5 Flash Lite medium | 0.5 | 0.7 | $0.440 | 0/1 | 86.9s | |
| #218 | Step 3.7 Flash low | Stepfun | 0.7 | 0.7 | $0.449 | 0/1 | 83.6s |
| #159 | Qwen3.7 Max none | Qwen | 1.0 | 0.8 | $0.451 | 1/1 | 80.3s |
| #279 | GLM 5.1 none | Z.ai | 1.0 | 0.6 | $0.451 | 1/1 | 107.2s |
| #372 | GLM 5V Turbo medium | Z.ai | 0.0 | 0.5 | $0.457 | 0/0 | 0ms |
| #17 | GPT-6.1 Sol low | OpenAI | 1.0 | 1.0 | $0.460 | 1/1 | 46.6s |