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

17 high-fit models

136 with cost/task

10192838475665$0.010$0.030$0.100$0.300$1.00$3.00Agentic IndexCost per taskGPT-5.6 Sol (xhigh) · Agentic 51.8 · $0.944 per taskClaude Opus 5 (Adaptive Reasoning, High Effort) · Agentic 52.1 · $1.23 per taskClaude Opus 5 (Adaptive Reasoning, Xhigh Effort) · Agentic 54.5 · $1.80 per taskGPT-5.6 Sol (high) · Agentic 48.5 · $0.771 per taskGPT-5.6 Sol (max) · Agentic 54 · $1.54 per taskClaude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) · Agentic 52.8 · $2.75 per taskClaude Opus 5 (Adaptive Reasoning, Max Effort) · Agentic 55.3 · $2.03 per taskKimi K3 (max) · Agentic 50.1 · $0.723 per taskClaude Opus 5 (Adaptive Reasoning, Medium Effort) · Agentic 47.1 · $0.618 per taskGPT-5.6 Sol (medium) · Agentic 44.5 · $0.514 per taskGPT-5.6 Terra (xhigh) · Agentic 44.7 · $0.430 per taskClaude Opus 4.8 (Adaptive Reasoning, Max Effort) · Agentic 47.2 · $2.03 per taskGPT-5.6 Terra (max) · Agentic 47.4 · $0.624 per taskGrok 4.5 (high) · Agentic 45.7 · $0.441 per taskGPT-5.5 (high) · Agentic 43.5 · $0.668 per taskGLM-5.2 (max) · Agentic 43.1 · $0.267 per taskGPT-5.5 (xhigh) · Agentic 44.9 · $1.17 per taskClaude Opus 4.7 (Adaptive Reasoning, Max Effort) · Agentic 44.4 · $1.97 per taskGPT-5.6 Luna (max) · Agentic 45.6 · $0.066 per taskGPT-5.6 Terra (high) · Agentic 41.3 · $0.304 per taskGPT-5.6 Sol (low) · Agentic 40 · $0.307 per taskGPT-5.6 Luna (xhigh) · Agentic 42.9 · $0.043 per taskClaude Sonnet 5 (Adaptive Reasoning, Max Effort) · Agentic 46.7 · $1.72 per taskGPT-5.5 (medium) · Agentic 37.8 · $0.495 per taskClaude Opus 5 (Adaptive Reasoning, Low Effort) · Agentic 39.8 · $0.426 per taskGPT-5.6 Luna (high) · Agentic 40.1 · $0.025 per taskGPT-5.4 (xhigh) · Agentic 41.1 · $0.931 per taskGemini 3.6 Flash (high) · Agentic 38.7 · $0.557 per taskGemini 3.5 Flash (high) · Agentic 37.4 · $0.691 per taskMuse Spark 1.1 (xhigh) · Agentic 37.5 · $0.261 per taskGPT-5.6 Terra (medium) · Agentic 37 · $0.160 per taskClaude Sonnet 4.6 (Adaptive Reasoning, Max Effort) · Agentic 40.8 · $1.14 per taskKimi K3 (low) · Agentic 37 · $0.243 per taskDeepSeek V4 Pro (Reasoning, Max Effort) · Agentic 36.4 · $0.045 per taskMiniMax-M3 · Agentic 35.4 · $0.137 per taskGPT-5.6 Sol (Non-reasoning) · Agentic 34.9 · $0.320 per taskClaude Sonnet 5 (Non-reasoning, High Effort) · Agentic 33.7 · $0.374 per taskDeepSeek V4 Pro (Reasoning, High Effort) · Agentic 34.4 · $0.043 per taskGPT-5.5 (low) · Agentic 30.4 · $0.211 per taskQwen3.7 Max · Agentic 30.6 · $1.28 per taskGPT-5.6 Terra (low) · Agentic 30.6 · $0.116 per taskGPT-5.4 mini (xhigh) · Agentic 30.2 · $0.452 per taskHy3 · Agentic 30.7 · $0.033 per taskDeepSeek V4 Flash (Reasoning, Max Effort) · Agentic 31.1 · $0.022 per taskMiMo-V2.5-Pro · Agentic 29.1 · $0.047 per taskKimi K2.7 Code · Agentic 29.6 · $0.221 per taskGPT-5.6 Luna (medium) · Agentic 31 · $0.015 per taskInkling Small · Agentic 30.8 · $0.073 per taskKimi K2.6 · Agentic 30.3 · $0.365 per taskGLM-5.1 (Reasoning) · Agentic 29.9 · $0.273 per taskGPT-5.4 nano (xhigh) · Agentic 27.5 · $0.133 per taskGPT-5.6 Terra (Non-reasoning) · Agentic 29.3 · $0.151 per taskGemini 3.1 Pro Preview · Agentic 21.4 · $0.291 per taskGrok Build 0.1 0616 · Agentic 28 · $0.223 per taskGPT-5.5 (Non-reasoning) · Agentic 25.8 · $0.206 per taskDeepSeek V4 Flash (Reasoning, High Effort) · Agentic 28.2 · $0.050 per taskNemotron 3 Ultra 550B A55B (Reasoning) · Agentic 27.4 · $0.413 per taskQwen3.6 Plus · Agentic 27.6 · $0.314 per taskGemini 3.5 Flash-Lite · Agentic 26.8 · $0.096 per taskMiMo-V2.5 · Agentic 23.7 · $0.010 per taskQwen3.6 27B (Reasoning) · Agentic 27 · $0.293 per taskClaude 4.5 Sonnet (Reasoning) · Agentic 24.6 · $0.462 per taskGPT-5.6 Luna (low) · Agentic 25.4 · $0.013 per taskQwen3.7 Plus · Agentic 20.8 · $0.226 per taskGLM-4.7 (Reasoning) · Agentic 25.4 · $0.353 per taskGrok 4.3 (high) · Agentic 24.1 · $0.145 per taskGPT-5.1 (high) · Agentic 21 · $0.300 per taskQwen3.6 27B (Non-reasoning) · Agentic 23.3 · $0.396 per taskGPT-5 (high) · Agentic 25.7 · $0.257 per taskStep 3.7 Flash · Agentic 21.5 · $0.091 per taskQwen3.5 122B A10B (Reasoning) · Agentic 20.7 · $0.257 per taskQwen3.5 397B A17B (Reasoning) · Agentic 19.8 · $0.359 per taskLongCat 2.0 · Agentic 21.8 · $0.122 per taskQwen3.6 35B A3B (Reasoning) · Agentic 21.4 · $0.194 per taskGPT-5.6 Luna (Non-reasoning) · Agentic 22 · $0.017 per taskMistral Medium 3.5 · Agentic 19 · $0.557 per taskRing-2.6-1T · Agentic 18.9 · $0.345 per taskGrok 4.3 (Non-reasoning) · Agentic 22.8 · $0.294 per taskQwen3.5 122B A10B (Non-reasoning) · Agentic 15.8 · $0.192 per taskGLM-4.6 (Reasoning) · Agentic 17.7 · $0.302 per taskGPT-5.6 Sol (xhigh)Claude Opus 5 (High EfClaude Opus 5 (Xhigh EGPT-5.6 Sol (high)GPT-5.6 Sol (max)

GPT-5.6 Sol (xhigh)

OpenAI

Agentic

51.8

Task cost

$0.944

ModelFitAgenticTask costResponseContext
GPT-5.6 Sol (xhigh)

OpenAI · Proprietary

9251.8$0.94453.2s1.0M
Claude Opus 5 (High Effort)

Anthropic · Proprietary

8852.1$1.2324.7s1.0M
Claude Opus 5 (Xhigh Effort)

Anthropic · Proprietary

8854.5$1.8034.3s1.0M
GPT-5.6 Sol (high)

OpenAI · Proprietary

8748.5$0.77121.7s1.0M
GPT-5.6 Sol (max)

OpenAI · Proprietary

8754$1.542.6m1.0M
Claude Fable 5 (Max Effort, Opus 4.8 Fallback)

Anthropic · Proprietary

8552.8$2.751.6m1.0M
Claude Opus 5 (max)

Anthropic · Proprietary

8455.3$2.031.5m1.0M
Kimi K3 (max)

Kimi · Open

8350.1$0.7231.3m1.0M
Claude Opus 5 (Medium Effort)

Anthropic · Proprietary

8047.1$0.61815.6s1.0M
GPT-5.6 Sol (medium)

OpenAI · Proprietary

8044.5$0.51413.6s1.0M

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.