Model comparison

Claude Opus 5 vs GPT-6 Astra

Anthropic against OpenAI, compared on context, price, and verified benchmark results.

Catalog record checked September 5, 2026Individual provider fields may changeEvidence confidence: High — see receipts below

Anthropic

Claude Opus 5

Frontier

OpenAI

GPT-6 Astra

Frontier

AI model capability comparison
Specification	Claude Opus 5	GPT-6 Astra
Provider	Anthropic	OpenAI
Tier	Frontier	Frontier
Context window	1M	Winner: 1.05M
Max output	128K	128K
Input / 1M tokens	Winner: $5	$10
Output / 1M tokens	Winner: $25	$50
Weights	Closed	Closed
Parameters	Not disclosedUnverified	computer-use flagship
Reasoning levels	adaptive	low, medium, high, xhigh, max
Modalities	text, image	text, image
API model id	claude-opus-5	gpt-6-astra
Released	July 24, 2026	September 3, 2026
Artificial Analysis Intelligence Index [max] (2026-08-14)	Winner: 63	61.2
Humanity's Last Exam (2026-09-01)	Winner: 63.6	57.2
Terminal-Bench 2.1 (2026-09-03)	86.7	Winner: 89.9
DeepSWE 1.1 (2026-09-03)	Not verifiedUnverified	74.1
OSWorld 2.0 (offline) (2026-09-03)	Not verifiedUnverified	72.6
GPQA Diamond (2026-09-03)	Not verifiedUnverified	96

Prices are USD per million tokens at standard rates, excluding batch and caching discounts. Bold indicates the better figure where one is objectively better. Values we could not confirm from the provider are shown as “Not verified” rather than estimated.

Benchmark receipts

2026-09-05: Artificial Analysis Terminal-Bench v2.1 (independent, Terminus 2 harness — conflicts with tbench.ai 87.4; harness difference preserved)Agentic terminal work: multi-step tasks executed in a sandboxed shell environment. Comparability: comparable with caveat — Scores across Terminal-Bench major versions (v1 vs v2) are NOT comparable; harness configuration (container, time limits) also shifts results.
2026-09-03: Terminal-Bench official leaderboard (independent, native harness, high effort; captured 2026-09-03 — the live board has since moved to Terminal-Bench 4.0)Agentic terminal work: multi-step tasks executed in a sandboxed shell environment. Comparability: comparable with caveat — Scores across Terminal-Bench major versions (v1 vs v2) are NOT comparable; harness configuration (container, time limits) also shifts results.
2026-09-03: OpenAI GPT-6 Astra launch chart (vendor, with tools; transcribed by Vellum)Agentic software engineering: pass@1 on real repo tasks with reported cost per completed task. Comparability: directly comparable
2026-09-03: Meta Muse Spark 1.3 launch table (independent cross-vendor run; conflicts with Google's 89.1 — both preserved)Agentic terminal work: multi-step tasks executed in a sandboxed shell environment. Comparability: comparable with caveat — Scores across Terminal-Bench major versions (v1 vs v2) are NOT comparable; harness configuration (container, time limits) also shifts results.
2026-09-03: Artificial AnalysisComposite index blending reasoning, knowledge, and coding evals into one 0–100 score. Comparability: directly comparable — AA occasionally rebaselines the index scale between snapshots — a score captured on one date is only comparable to same-snapshot scores (check measuredAt).
2026-09-02: Google Gemini 3.8 Flash launch eval table (independent cross-vendor run)Agentic terminal work: multi-step tasks executed in a sandboxed shell environment. Comparability: comparable with caveat — Scores across Terminal-Bench major versions (v1 vs v2) are NOT comparable; harness configuration (container, time limits) also shifts results.
2026-09-01: Anthropic Fable 5.1 comparison table (vendor, with tools; 56.6 no tools)Frontier-knowledge ceiling: extremely hard multi-domain questions written to be near-impossible without deep expertise. Comparability: comparable with caveat — Subset/tool configurations (text-only vs with-tools) differ between vendors and materially change scores; must match configuration to compare.

Pricing tiers: Claude Opus 5: Standard $5/$25 per MTok; cache reads 10% of base input. Knowledge cutoff May 2026. · GPT-6 Astra: Standard $10/$50 per MTok (≤272K context); >272K input billed $20/$75 (Fast mode output $100). Cached input $1/MTok; Fast mode is 2x price for ~2.5x speed. Knowledge cutoff Apr 30 2026. Rolling out over several days from 2026-09-03; Enterprise tenants get it disabled by default.

FrontierRecord checked September 3, 2026

Claude Opus 5

Claude Opus 5 is Anthropic's top model for complex agentic coding and enterprise work — superseded at the top of the Artificial Analysis Index by Fable 5.1 (2026-09-01) but stronger on price.

Best for

Complex agentic coding
Enterprise work
Long-context analysis

Watch out

Output costs 5x input; long generations dominate the bill.

FrontierRecord checked September 5, 2026

GPT-6 Astra

GPT-6 Astra is OpenAI's computer-use flagship — the first model it gates at the Preparedness Framework's 'Critical' cybersecurity threshold, trained on >100,000 GPUs at Stargate.

Best for

Computer-use agents
Agentic coding
Hard reasoning

Watch out

Gated at the 'Critical' cyber threshold and disabled by default for Enterprise; long-context work bills $20/$75 above 272K input tokens.

When the cheaper one wins

Claude Opus 5 is cheaper on output at $25 per million tokens against $50 for GPT-6 Astra — about 2.0×. Use the cheaper tier for classification, extraction, summarisation, and any task where the expensive model’s extra score does not change the accepted output. The expensive one only pays if your hardest task actually fails on the cheap tier. These are standard-tier API rates, excluding batch and cache discounts.

Run the model picker

Evidence confidence: High

How strong and complete the evidence behind this comparison is — not a prediction of which model is better.

Pricing verified on both sides — Input and output rates are verified for both models.
4/5 core specs verified on both sides — Not published for at least one side: parameter count.
4 shared named benchmarks with differing scores — Measured on: Artificial Analysis Intelligence Index, Humanity's Last Exam, Terminal-Bench 2.1, Terminal-Bench 2.1.
Verified within the last 90 days — Newest catalog check was 6 days ago.
Both models carry source citations — Each side has at least two catalog sources on record.

Source receipts

Each catalog figure was checked against the provider or an independent second source on the date shown.

Claude Opus 5: Anthropic — Claude Opus 5 (accessed 2026-08-29)
Claude Opus 5: Anthropic news — Claude Opus 5 (accessed 2026-08-29)
GPT-6 Astra: OpenAI — Introducing GPT-6 Astra (accessed 2026-09-03)
GPT-6 Astra: OpenAI API pricing (gpt-6-astra $10/$50) (accessed 2026-09-03)
GPT-6 Astra: CNBC — OpenAI launches GPT-6 Astra (accessed 2026-09-03)
GPT-6 Astra: OpenAI model docs — gpt-6-astra (128K max output, effort levels) (accessed 2026-09-05)

Related comparisons

Diving deeper on one model? Claude Opus 5 · GPT-6 Astra

Common questions

Claude Opus 5 vs GPT-6 Astra

Answered from the verified figures on this page rather than general guidance.

Is Claude Opus 5 or GPT-6 Astra cheaper for input?

Claude Opus 5 is cheaper at $5 per million input tokens, against $10 for GPT-6 Astra — roughly 2.0× the price. Output tokens usually dominate a real bill, so weigh the output rate more heavily than the input rate. These are base rates; Claude Opus 5 has tiered pricing: Standard $5/$25 per MTok; cache reads 10% of base input. Knowledge cutoff May 2026. GPT-6 Astra has tiered pricing: Standard $10/$50 per MTok (≤272K context); >272K input billed $20/$75 (Fast mode output $100). Cached input $1/MTok; Fast mode is 2x price for ~2.5x speed. Knowledge cutoff Apr 30 2026. Rolling out over several days from 2026-09-03; Enterprise tenants get it disabled by default.

Is Claude Opus 5 or GPT-6 Astra cheaper for output?

Claude Opus 5 is cheaper at $25 per million output tokens, against $50 for GPT-6 Astra — roughly 2.0× the price. Output tokens usually dominate a real bill, so weigh the output rate more heavily than the input rate. These are base rates; Claude Opus 5 has tiered pricing: Standard $5/$25 per MTok; cache reads 10% of base input. Knowledge cutoff May 2026. GPT-6 Astra has tiered pricing: Standard $10/$50 per MTok (≤272K context); >272K input billed $20/$75 (Fast mode output $100). Cached input $1/MTok; Fast mode is 2x price for ~2.5x speed. Knowledge cutoff Apr 30 2026. Rolling out over several days from 2026-09-03; Enterprise tenants get it disabled by default.

Which has the larger context window, Claude Opus 5 or GPT-6 Astra?

GPT-6 Astra accepts 1.05M tokens against 1M for Claude Opus 5. This only matters if you routinely send very long documents or large codebases.

Do Claude Opus 5 and GPT-6 Astra support the same reasoning levels?

Claude Opus 5 exposes adaptive, while GPT-6 Astra exposes low, medium, high, xhigh, max.

Should I use Claude Opus 5 or GPT-6 Astra?

Both sit in the frontier tier, so the choice usually comes down to price and context rather than capability. Claude Opus 5 suits complex agentic coding; GPT-6 Astra suits computer-use agents.

Next step

Choosing between them

The verified figures that separate this pair, computed from the catalog rather than restated boilerplate.

Input price: Claude Opus 5 costs $5 per 1M tokens versus $10 for GPT-6 Astra — a 2x difference at the headline tier.
Context: GPT-6 Astra takes 1.05M against 1M for Claude Opus 5 — only decisive if your prompts approach the smaller window.
Measured capability: Claude Opus 5 leads Artificial Analysis Intelligence Index 63 to 61.2 (measured 2026-08-14).

Run the model picker Estimate stack cost All model comparisons

Choosing a harness rather than a model — Cursor, Copilot, Claude Code, Windsurf, or Lovable? Compare agentic harnesses · Latest releases.