GPT/Flagship

GPT 6 Astra

GPT 6 Astra is GPT's most capable model, priced at $10.00/MTok input and $50.00/MTok output with a 922K context window. Backfilled on 2026-09-23. Prices confirmed by the provider pricing page and the LiteLLM cross-check. Release date not published in a machine-readable source. Not yet benchmarked, so never recommended.

Input $/MTok

$10.00

Output $/MTok

$50.00

Context window

922K

Cache read discount

90% off

GPT 6 Astra pricing breakdown

RateInput $/MTokOutput $/MTok
Standard$10.00$50.00
Cached read (90% off input)$1.00-
Cached write (25% surcharge)$12.50-
Batch (50% off)$5.00$25.00
Long context (≥272.001K tokens)$20.00$75.00

All rates in USD per million tokens. Prices verified against provider documentation.

GPT 6 Astra task benchmarks

Confidence scores (0-100) and cross-provider rank for each task type. Rank 1 is best across all 72 models.

Not benchmarked. The third-party benchmark source used for confidence scores does not cover GPT models yet. Scores will appear here once independent benchmarks are published.

GPT 6 Astra cost by use case

Estimated cost per API call using default token counts for each workload.

Use caseInput tokensOutput tokensCost / call
Code Generation2,0004,000$0.2200
Unit Test Generation3,0005,000$0.2800
SQL Generation1,5001,000$0.0650
API Generation3,0006,000$0.3300
Code Refactoring4,0004,000$0.2400
Code Review5,0002,000$0.1500
PR Review8,0003,000$0.2300
Security Review6,0002,500$0.1850
Bug Detection5,0002,000$0.1500
Document Summarization10,0001,000$0.1500
PDF Summarization15,0001,500$0.2250
Meeting Notes8,0002,000$0.1800

Monthly cost projections

Projected monthly spend at different call volumes, using code generation as a representative workload (2,000 input / 4,000 output tokens per call).

Calls / monthWithout cachingWith caching (90% read discount)
1K$220$207.4
10K$2,200$2,074
100K$22,000$20,740
500K$110,000$103,700
1M$220,000$207,400

Caching estimate assumes 70% of input tokens are cache reads. Your actual cache hit rate will depend on prompt structure and reuse.

Flagship alternatives to GPT 6 Astra

Same-tier models from other providers, sorted by input price.

ModelProviderInput $/MTokOutput $/MTokInput savings
Grok 4.6Grok$2.00$6.0080% cheaper
Grok 4.5Grok$2.00$6.0080% cheaper
Grok 4.7Grok$2.00$6.0080% cheaper
Claude Opus 5.5Claude$4.00$20.0060% cheaper
Claude Opus 4.8Claude$5.00$25.0050% cheaper
Claude Opus 5Claude$5.00$25.0050% cheaper
Claude Opus 4.7Claude$5.00$25.0050% cheaper
Claude Opus 4.6Claude$5.00$25.0050% cheaper
Claude Opus 4.5Claude$5.00$25.0050% cheaper
Claude Fable 5.1Claude$10.00$50.00Same price
Claude Fable 5Claude$10.00$50.00Same price
Claude Mythos 5.1Claude$10.00$50.00Same price
Claude Mythos 5Claude$10.00$50.00Same price

Other GPT models

Track your GPT 6 Astra spend

See exactly how much you spend on GPT 6 Astra and where you can save. Two-line wrapper install, no API keys shared.