GPT-6 Astra vs Claude Opus 5

GPT-6 Astra by OpenAI · Claude Opus 5 by Anthropic · 29 shared benchmarks

GPT-6 Astra

77.6

#3 · 100%

Claude Opus 5

74.9

#5 · 100%

SI Score, rank and confidence side by side

Pillars — GPT-6 Astra

Coding 64.8 / 69.7

Math 92.9 / 86.9

Preference 77.1 / 81.9

Reasoning 87.1 / 84.3

Right-hand value is Claude Opus 5.

The basics

Attribute	GPT-6 Astra	Claude Opus 5
Input price / 1M	$10.00	$5.00
Output price / 1M	$50.00	$25.00
Context window	1.1M	1M
Released	Sep 3, 2026	Jul 24, 2026
Open weights	Closed	Closed

Shared benchmarks

29 in common

Only benchmarks both models have are compared. Raw values are in each benchmark's own unit; a direct comparison requires matching evaluation conditions. All reported source/variant rows are shown; no raw-result winner is assigned across unmatched harnesses. Sources and dates sit behind every dotted number.

Which should you choose?

What the data says — heuristics from the numbers above, not a verdict:

For math, GPT-6 Astra leads by 6.0 normalized points.
For coding, Claude Opus 5 leads by 4.9 normalized points.
For preference, Claude Opus 5 leads by 4.9 normalized points.
Claude Opus 5 has the lower input-token price ($5.00 vs $10.00 per 1M).
Confidence differs: GPT-6 Astra 100% vs Claude Opus 5 100% — pending sources can still move either score.

Benchmark	GPT-6 Astra	Claude Opus 5	Conditions
arc agi 3	99.9%Lab-reported; metric RHAE; transcribed by MIT models.dev catalog; not independently evaluated [variant] Responses API	30.2%Lab-reported; metric RHAE; transcribed by MIT models.dev catalog; not independently evaluated [variant] high effort	Check variant and harness
arc agi v1 public eval	98.5%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'high' reasoning effort. [variant] openai-gpt-6-astra-high 98.3%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'low' reasoning effort. [variant] openai-gpt-6-astra-low 97.8%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'max' reasoning effort. [variant] openai-gpt-6-astra-max 98.8%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'medium' reasoning effort. [variant] openai-gpt-6-astra-medium 99%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'xhigh' reasoning effort. [variant] openai-gpt-6-astra-xhigh	99%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'high' output effort. [variant] anthropic-claude-opus-5-high 99%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'max' output effort. [variant] anthropic-claude-opus-5-max	Check variant and harness
arc agi v1 semi private	98.5%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'high' reasoning effort. [variant] openai-gpt-6-astra-high 96.5%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'low' reasoning effort. [variant] openai-gpt-6-astra-low 97.5%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'max' reasoning effort. [variant] openai-gpt-6-astra-max 97.5%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'medium' reasoning effort. [variant] openai-gpt-6-astra-medium 98.5%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'xhigh' reasoning effort. [variant] openai-gpt-6-astra-xhigh	97.5%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'high' output effort. [variant] anthropic-claude-opus-5-high 97.5%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'max' output effort. [variant] anthropic-claude-opus-5-max	Check variant and harness
arc agi v2 public eval	97.9%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'high' reasoning effort. [variant] openai-gpt-6-astra-high 94.2%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'low' reasoning effort. [variant] openai-gpt-6-astra-low 97.5%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'max' reasoning effort. [variant] openai-gpt-6-astra-max 96.7%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'medium' reasoning effort. [variant] openai-gpt-6-astra-medium 97.9%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'xhigh' reasoning effort. [variant] openai-gpt-6-astra-xhigh	93.2%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'high' output effort. [variant] anthropic-claude-opus-5-high 97.1%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'max' output effort. [variant] anthropic-claude-opus-5-max	Check variant and harness
arc agi v2 semi private	92.1%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'high' reasoning effort. [variant] openai-gpt-6-astra-high 85.4%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'low' reasoning effort. [variant] openai-gpt-6-astra-low 95%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'max' reasoning effort. [variant] openai-gpt-6-astra-max 92.1%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'medium' reasoning effort. [variant] openai-gpt-6-astra-medium 93.3%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'xhigh' reasoning effort. [variant] openai-gpt-6-astra-xhigh	88.3%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'high' output effort. [variant] anthropic-claude-opus-5-high 90.4%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'max' output effort. [variant] anthropic-claude-opus-5-max	Check variant and harness
arc agi v3 semi private	54.8%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'high' reasoning effort. [variant] openai-gpt-6-astra-high 17.5%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'low' reasoning effort. [variant] openai-gpt-6-astra-low 62.7%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'max' reasoning effort. [variant] openai-gpt-6-astra-max 38.6%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'medium' reasoning effort. [variant] openai-gpt-6-astra-medium 59.3%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'xhigh' reasoning effort. [variant] openai-gpt-6-astra-xhigh	30.2%ARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with 'high' output effort. [variant] anthropic-claude-opus-5-high	Check variant and harness
frontiermath tier 4 v2	97.6%Epoch-owned evaluation mean score; scores only, no benchmark questions [variant] high 87.8%Epoch-owned evaluation mean score; scores only, no benchmark questions [variant] low 97.6%Epoch-owned evaluation mean score; scores only, no benchmark questions [variant] max 97.6%Epoch-owned evaluation mean score; scores only, no benchmark questions [variant] medium 82.9%Epoch-owned evaluation mean score; scores only, no benchmark questions [variant] none 97.6%Epoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhigh	73.2%Epoch-owned evaluation mean score; scores only, no benchmark questions [variant] max	Check variant and harness
frontiermath tiers 1 3 v2	93.7%Epoch-owned evaluation mean score; scores only, no benchmark questions [variant] max	85.6%Epoch-owned evaluation mean score; scores only, no benchmark questions [variant] max	Check variant and harness
gpqa diamond	95.8%Epoch-owned evaluation mean score; scores only, no benchmark questions [variant] max 96%Lab-reported; metric accuracy; transcribed by MIT models.dev catalog; not independently evaluated [variant]	92.9%Epoch-owned evaluation mean score; scores only, no benchmark questions [variant] 87.9%Epoch-owned evaluation mean score; scores only, no benchmark questions [variant] low 93.9%Epoch-owned evaluation mean score; scores only, no benchmark questions [variant] max	Check variant and harness
hle tools	57.2%Lab-reported; metric accuracy; transcribed by MIT models.dev catalog; not independently evaluated [variant] with tools	64.7%Lab-reported; metric accuracy; transcribed by MIT models.dev catalog; not independently evaluated [variant] with tools	Check variant and harness
livebench coding code completion 2026_06_25	80.4%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	82.6%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	Check variant and harness
livebench coding code generation 2026_06_25	80.3%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	80.3%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	Check variant and harness
livebench coding javascript 2026_06_25	63.6%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	77.3%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	Check variant and harness
livebench coding python 2026_06_25	65%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	75%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	Check variant and harness
livebench coding typescript 2026_06_25	43.3%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	43.3%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	Check variant and harness
livebench math amps hard 2026_06_25	98%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	99.0%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	Check variant and harness
livebench math integrals with game 2026_06_25	100%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	97%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	Check variant and harness
livebench math math comp 2026_06_25	97.1%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	94.1%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	Check variant and harness
livebench math olympiad 2026_06_25	92.2%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	92.8%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	Check variant and harness
livebench math simplify 2026_06_25	70.5%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	61.6%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	Check variant and harness
livebench reasoning connections 2026_06_25	100%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	99.3%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	Check variant and harness
livebench reasoning consecutive events 2026_06_25	90.6%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	77.6%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	Check variant and harness
livebench reasoning logic with navigation 2026_06_25	88%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	86%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	Check variant and harness
livebench reasoning spatial 2026_06_25	98%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	100%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	Check variant and harness
livebench reasoning theory of mind 2026_06_25	84.6%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	78.8%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	Check variant and harness
livebench reasoning zebra puzzle 2026_06_25	100%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	100%LiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25	Check variant and harness
lmarena text	1442.3 eloPublished source fact [variant] text / overall	1502.1 eloPublished source fact [variant] text / overall	Check variant and harness
otis mock aime 2024 2025	100%Epoch-owned evaluation mean score; scores only, no benchmark questions [variant] max	97.8%Epoch-owned evaluation mean score; scores only, no benchmark questions [variant] 93.3%Epoch-owned evaluation mean score; scores only, no benchmark questions [variant] low 98.9%Epoch-owned evaluation mean score; scores only, no benchmark questions [variant] max	Check variant and harness
terminal bench v4 0	57.9%Lab-reported; metric success rate; transcribed by MIT models.dev catalog; not independently evaluated [variant] 4.0 57.9%Terminal-Bench 4.0; published harness submission, 95% CI retained at source [variant] Codex; high 50.6%Terminal-Bench 4.0; published harness submission, 95% CI retained at source [variant] Codex; low 58.2%Terminal-Bench 4.0; published harness submission, 95% CI retained at source [variant] Codex; max 54.2%Terminal-Bench 4.0; published harness submission, 95% CI retained at source [variant] Codex; medium 57.9%Terminal-Bench 4.0; published harness submission, 95% CI retained at source [variant] Codex; xhigh	50.3%Terminal-Bench 4.0; published harness submission, 95% CI retained at source [variant] Claude Code; high 34.9%Terminal-Bench 4.0; published harness submission, 95% CI retained at source [variant] Claude Code; low 51.8%Terminal-Bench 4.0; published harness submission, 95% CI retained at source [variant] Claude Code; max 44.9%Terminal-Bench 4.0; published harness submission, 95% CI retained at source [variant] Claude Code; medium 53.9%Terminal-Bench 4.0; published harness submission, 95% CI retained at source [variant] Claude Code; xhigh	Check variant and harness