Agentic model routing

Find the right model for agent work

Match a workload to models using Agentic Index, task-level cost, response time, benchmark signals, and context requirements. The shortlist is built from the same live Artificial Analysis data used across WhatLLM.

Agentic fit lab

Choose the model that fits the work

Ranking changes as task shape, autonomy, context, and economics change.

Task

Autonomy needed

Context load

Optimize for

Frontier map

Agentic Index vs task cost

14 high-fit models

112 with cost/task

0112233435465$0.010$0.030$0.100$0.300$1.00$3.00Agentic IndexCost per taskGPT-5.6 Sol (xhigh) · Agentic 51.8 · $0.682 per taskGPT-5.6 Sol (max) · Agentic 54 · $1.04 per taskGPT-5.6 Sol (high) · Agentic 48.5 · $0.453 per taskClaude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) · Agentic 52.8 · $2.75 per taskKimi K3 · Agentic 50.1 · $0.954 per taskGPT-5.6 Sol (medium) · Agentic 44.5 · $0.314 per taskGPT-5.6 Terra (xhigh) · Agentic 44.7 · $0.477 per taskClaude Opus 4.8 (Adaptive Reasoning, Max Effort) · Agentic 47.2 · $1.80 per taskGrok 4.5 (high) · Agentic 45.7 · $0.312 per taskGPT-5.6 Terra (max) · Agentic 47.4 · $0.825 per taskGPT-5.5 (xhigh) · Agentic 44.9 · $0.993 per taskGPT-5.5 (high) · Agentic 43.5 · $0.668 per taskGLM-5.2 (max) · Agentic 43.1 · $0.468 per taskGPT-5.6 Luna (max) · Agentic 45.6 · $0.209 per taskClaude Opus 4.7 (Adaptive Reasoning, Max Effort) · Agentic 44.4 · $1.97 per taskGPT-5.6 Terra (high) · Agentic 41.3 · $0.336 per taskGPT-5.6 Sol (low) · Agentic 40 · $0.198 per taskGPT-5.6 Luna (xhigh) · Agentic 42.9 · $0.139 per taskClaude Sonnet 5 (Adaptive Reasoning, Max Effort) · Agentic 46.7 · $1.53 per taskGPT-5.5 (medium) · Agentic 37.8 · $0.406 per taskGPT-5.6 Luna (high) · Agentic 40.1 · $0.095 per taskGemini 3.5 Flash (high) · Agentic 37.4 · $0.586 per taskMuse Spark 1.1 (xhigh) · Agentic 37.5 · $0.261 per taskClaude Sonnet 4.6 (Adaptive Reasoning, Max Effort) · Agentic 40.8 · $1.14 per taskGPT-5.6 Terra (medium) · Agentic 37 · $0.175 per taskDeepSeek V4 Pro (Reasoning, Max Effort) · Agentic 36.4 · $0.045 per taskMiniMax-M3 · Agentic 35.4 · $0.125 per taskGPT-5.6 Sol (Non-reasoning) · Agentic 34.9 · $0.200 per taskDeepSeek V4 Pro (Reasoning, High Effort) · Agentic 34.4 · $0.041 per taskClaude Sonnet 5 (Non-reasoning, High Effort) · Agentic 33.7 · $0.374 per taskQwen3.7 Max · Agentic 30.6 · $1.03 per taskGPT-5.5 (low) · Agentic 30.4 · $0.211 per taskGPT-5.6 Terra (low) · Agentic 30.6 · $0.154 per taskGPT-5.4 mini (xhigh) · Agentic 30.2 · $0.452 per taskMiMo-V2.5-Pro · Agentic 29.1 · $0.031 per taskDeepSeek V4 Flash (Reasoning, Max Effort) · Agentic 31.1 · $0.022 per taskKimi K2.6 · Agentic 30.3 · $0.335 per taskGLM-5.1 (Reasoning) · Agentic 29.9 · $0.226 per taskGPT-5.6 Luna (medium) · Agentic 31 · $0.050 per taskGPT-5.4 nano (xhigh) · Agentic 27.5 · $0.133 per taskGPT-5.6 Terra (Non-reasoning) · Agentic 29.3 · $0.179 per taskGemini 3.1 Pro Preview · Agentic 21.4 · $0.291 per taskGPT-5.5 (Non-reasoning) · Agentic 25.8 · $0.171 per taskGrok Build 0.1 0616 · Agentic 28 · $0.212 per taskNemotron 3 Ultra 550B A55B (Reasoning) · Agentic 27.4 · $0.244 per taskQwen3.6 Plus · Agentic 27.6 · $0.314 per taskMiMo-V2.5 · Agentic 23.7 · $0.010 per taskQwen3.6 27B (Reasoning) · Agentic 27 · $0.265 per taskClaude 4.5 Sonnet (Reasoning) · Agentic 24.6 · $0.413 per taskGPT-5.6 Luna (low) · Agentic 25.4 · $0.041 per taskQwen3.7 Plus · Agentic 20.8 · $0.207 per taskGLM-4.7 (Reasoning) · Agentic 25.4 · $0.323 per taskGrok 4.3 (high) · Agentic 24.1 · $0.139 per taskGPT-5.1 (high) · Agentic 21 · $0.270 per taskQwen3.6 27B (Non-reasoning) · Agentic 23.3 · $0.359 per taskGPT-5 (high) · Agentic 25.7 · $0.238 per taskQwen3.5 122B A10B (Reasoning) · Agentic 20.7 · $0.241 per taskQwen3.5 397B A17B (Reasoning) · Agentic 19.8 · $0.333 per taskQwen3.6 35B A3B (Reasoning) · Agentic 21.4 · $0.179 per taskGPT-5.6 Luna (Non-reasoning) · Agentic 22 · $0.055 per taskMistral Medium 3.5 · Agentic 19 · $0.563 per taskRing-2.6-1T · Agentic 18.9 · $0.345 per taskGrok 4.3 (Non-reasoning) · Agentic 22.8 · $0.293 per taskQwen3.5 122B A10B (Non-reasoning) · Agentic 15.8 · $0.177 per taskGLM-4.6 (Reasoning) · Agentic 17.7 · $0.283 per taskClaude 4.5 Haiku (Reasoning) · Agentic 16.4 · $0.237 per taskMercury 2 · Agentic 9.6 · $0.075 per taskMistral Medium 3.1 · Agentic 6.2 · $0.145 per taskNova 2.0 Pro Preview (medium) · Agentic 7 · $0.173 per taskDeepSeek R1 (Jan '25) · Agentic 3.1 · $0.247 per taskMistral Small 3.1 · Agentic 5.2 · $0.041 per taskGemini 2.5 Pro · Agentic 7.1 · $0.198 per taskQwen3.5 9B (Reasoning) · Agentic 7.4 · $0.164 per taskHyperNova 60B 2605 · Agentic 6.7 · $0.018 per taskgpt-oss-20b (high) · Agentic 3.1 · $0.018 per taskMistral Small 4 (Reasoning) · Agentic 4.7 · $0.098 per taskMistral Small 3.2 · Agentic 2 · $0.124 per taskDeepSeek V3 (Dec '24) · Agentic 1.6 · $0.023 per taskMistral Large 3 · Agentic 5.5 · $0.060 per taskGemini 3.1 Flash-Lite · Agentic 6.2 · $0.042 per taskGPT-5.6 Sol (xhigh)GPT-5.6 Sol (max)GPT-5.6 Sol (high)Claude Fable 5 (Max EfKimi K3

GPT-5.6 Sol (xhigh)

OpenAI

Agentic

51.8

Task cost

$0.682

ModelFitAgenticTask costResponseContext
GPT-5.6 Sol (xhigh)

OpenAI · Proprietary

9251.8$0.6821.0m1.0M
GPT-5.6 Sol (max)

OpenAI · Proprietary

8954$1.042.6m1.0M
GPT-5.6 Sol (high)

OpenAI · Proprietary

8848.5$0.45320.2s1.0M
Claude Fable 5 (Max Effort, Opus 4.8 Fallback)

Anthropic · Proprietary

8452.8$2.751.9m1.0M
Kimi K3

Kimi · Proprietary

8250.1$0.9541.1m1.0M
GPT-5.6 Sol (medium)

OpenAI · Proprietary

8044.5$0.31418.7s1.0M
GPT-5.6 Terra (xhigh)

OpenAI · Proprietary

8044.7$0.47714.3s1.0M
Claude Opus 4.8 (max)

Anthropic · Proprietary

7947.2$1.8039.3s1.0M
Grok 4.5 (high)

SpaceXAI · Proprietary

7745.7$0.31217.4s500K
GPT-5.6 Terra (max)

OpenAI · Proprietary

7647.4$0.8252.7m1.0M

Why cost per task changes the decision

Agent loops compound price and latency

A small per-token price gap can become a large bill when an agent runs many turns, calls tools, and carries long state. Cost per Intelligence Index task gives a cleaner decision unit than token price alone.

The best model depends on the failure cost

Frontier models are worth it when mistakes are expensive. For routine automation, a cheaper high-fit model can preserve most of the capability while cutting task cost sharply.