Comparing 2 models

Claude Sonnet 5.5 vs GPT-6 Astra

Benchmarks through

Verdict

Claude Sonnet 5.5 places higher on 5 of the 9 tasks where both are ranked. Claude Sonnet 5.5 costs 5.0x less per input token. GPT-6 Astra accepts 1.1x more context.

  • Choose Claude Sonnet 5.5 for Debugging & Code Review (#1), AI Agents (#4), Autonomous Agents (#4).
  • Choose GPT-6 Astra for Tool & Function Calling (#2), RAG & Long Context (#2), Math & Reasoning (#5).

Built from the published task rankings below; ranks come from each task's own weighted benchmarks, so a win on one task says nothing about another.

metric
Model 1Claude Sonnet 5.5
Model 2GPT-6 Astra
Best for
Code Completion (#1)
Grounded RAG (#2)
Provider
anthropic
openai
Context
1M
1.1MBest
Input / 1M
$2Best
$10
Output / 1M
$10Best
$50
Tool use
yesBest
yesBest
Vision
yesBest
yesBest
Reasoning
yesBest
yesBest
Open-weight
—
—
Released
Sep 28, 2026
Sep 4, 2026

Measured head to head

Benchmarks both models were run on

EvaluationClaude Sonnet 5.5GPT-6 AstraSource
FACTS Grounding62.5%80.1%source · 2026-11-12
FrontierMath Tiers 1-3 (Epoch AI)88.8% ± 1.993.7% ± 1.4source · 2026-09-29
GPQA Diamond (Epoch AI)95.6% ± 1.495.8% ± 1.4source · 2026-09-29
LiveBench Agentic Coding56.3%57.3%source · 2026-10-18
LiveBench Coding91.4%80.4%source · 2026-10-18
LiveBench Data Analysis59.5%83.0%source · 2026-10-18
LiveBench Instruction Following56.8%75.6%source · 2026-10-18
LiveBench Language78.0%89.4%source · 2026-10-18
LiveBench Math96.1%96.8%source · 2026-10-18
LiveBench Reasoning91.6%92.7%source · 2026-10-18
LMArena Agent1614.5source · 2026-11-11
LMArena Agent Bash Recovery155.5source · 2026-10-08
LMArena Agent Steerability14.514.8source · 2026-11-11
LMArena Agent Task Outcome19.514source · 2026-11-11
LMArena Agent Tool Hallucination0.91source · 2026-11-11
LMArena Creative Writing1466 Elo1421 Elosource · 2026-11-11
LMArena Instruction Following1487 Elo1456 Elosource · 2026-11-11
LMArena Long Query1487 Elo1459 Elosource · 2026-11-11
LMArena Math1483 Elo1459 Elosource · 2026-11-11
LMArena Text1473 Elo1474 Elosource · 2026-11-11
LMArena Text Factuality1472 Elo1466 Elosource · 2026-11-11
LMArena Vision1263 Elo1283 Elosource · 2026-11-09
LMArena Vision Diagrams1308 Elo1311 Elosource · 2026-11-09
LMArena Vision OCR1281 Elo1297 Elosource · 2026-11-09
LMArena WebDev1772 Elo1764 Elosource · 2026-11-11
LMArena WebDev Image-to-Web1744 Elo1722 Elosource · 2026-11-10
OTIS Mock AIME (Epoch AI)100% ± 0100% ± 0source · 2026-09-29
SimpleQA Verified (Epoch AI)46.5% ± 1.675.6% ± 1.4source · 2026-09-29
Terminal-Bench (4.1)63.9%60.3%source · 2026-11-15

Values are as published by each evaluator, in its own units; rows from different evaluators are not on a common scale.