GPT-6.2 AstravsGPT-6 Astra

They share the same context, max output, input modalities, and output modalities, and differ in input price ($8.00 / 1M tokens vs $10.00 / 1M tokens), output price ($40.00 / 1M tokens vs $50.00 / 1M tokens), and knowledge cutoff (May 2026 vs Apr 2026).

Compare at your usage

Every model priced for the same request
Prompt per request
Cache hits
Batch
Service tier

8,000 tokens in, 2,000 out

GPT-6.2 Astra on OpenAI APIGPT-6 Astra on OpenAI API
Rate appliedBase rateBase rate
Inputper 1M tokens$8.00 $10.00
Cache readper 1M tokens$0.40 $1.00
Outputper 1M tokens$40.00 $50.00
Cost per request$0.144$0.18
Per 1,000 requests$144.00$180.00

Bold marks the cheapest for this request at each maker’s own list prices.

Cost per request as prompts grow

2K tokens out, list price on each maker’s API. GPT-6.2 Astra charges $16.00 / $60.00 above 272K tokens. GPT-6 Astra charges $20.00 / $75.00 above 272K tokens.

0$5.00$10.00$15.000250K500K750K1Mprompt tokens per request272K8KGPT-6.2 Astra$0.144 at 8KGPT-6 Astra$0.18 at 8K
  • GPT-6.2 Astra · OpenAI API
  • GPT-6 Astra · OpenAI API

Key differences

InputMaker’s official API · global · USD
$8.00 / 1M tokens OpenAI API pricing · as of Nov 18, 2026
$10.00 / 1M tokens OpenAI API pricing · as of Nov 9, 2026
OutputMaker’s official API · global · USD
$40.00 / 1M tokens OpenAI API pricing · as of Nov 18, 2026
$50.00 / 1M tokens OpenAI API pricing · as of Nov 9, 2026
Knowledge cutoff
May 2026 GPT-6.2 Astra model · as of Nov 17, 2026
Apr 2026 GPT-6 Astra model · as of Nov 5, 2026
Cache readMaker’s official API · global · USD
$0.40 / 1M tokens OpenAI API pricing · as of Nov 18, 2026
$1.00 / 1M tokens OpenAI API pricing · as of Nov 9, 2026
Cache writeMaker’s official API · global · USD · 30-minute cache
$10.00 / 1M tokens OpenAI API pricing · as of Nov 18, 2026
$12.50 / 1M tokens OpenAI API pricing · as of Nov 9, 2026
Cache writeMaker’s official API · global · USD · 12-hour cache
$20.00 / 1M tokens OpenAI API pricing · as of Nov 18, 2026
$25.00 / 1M tokens OpenAI API pricing · as of Nov 9, 2026

Full comparison

GPT-6.2 Astra OpenAI GPT-6 Astra OpenAI
Overview
ReleasedNov 17, 2026 Sep 3, 2026
Open weights— —
Knowledge cutoffMay 2026 Apr 2026
Specs
Context1.05M 1.05M
Max input— 922K
Max output128K 128K
Input modalitiesText, Image Text, Image
Output modalitiesText Text
ReasoningYes Yes
Reasoning efforts— low, medium, high, xhigh, max
Reasoning can be off— —
Tool useYes Yes
Structured outputYes Yes
Pricing · per 1M tokens
Input price$8.00Also priced under other conditions: Prompts over 272K tokens, Batch, Fast tier, Ultrafast tier $10.00Also priced under other conditions: Batch, Flex tier, Fast tier, Prompts over 272K tokens, Ultrafast tier
Output price$40.00Also priced under other conditions: Prompts over 272K tokens, Batch, Fast tier, Ultrafast tier $50.00Also priced under other conditions: Batch, Flex tier, Fast tier, Prompts over 272K tokens, Ultrafast tier
Cache read price$0.40Also priced under other conditions: Prompts over 272K tokens, Batch, Fast tier, Ultrafast tier $1.00Also priced under other conditions: Batch, Flex tier, Fast tier, Prompts over 272K tokens, Ultrafast tier
Cache write price$10.00Also priced under other conditions: 30-minute cache, 12-hour cache, Prompts over 272K tokens, Batch, Fast tier, Ultrafast tier $12.50Also priced under other conditions: 30-minute cache, 12-hour cache, Prompts over 272K tokens, Batch, Flex tier, Fast tier, Ultrafast tier
Blended price (3 input : 1 output)$16.00 $20.00
Pricing under conditions · per 1M tokens
Input price, long prompts$16.00Prompts over 272K tokens $20.00Prompts over 272K tokens
Output price, long prompts$60.00Prompts over 272K tokens $75.00Prompts over 272K tokens
Batch input price$4.00Batch $5.00Batch
Batch output price$20.00Batch $25.00Batch
Flex tier input priceNot offered $5.00Flex tier
Flex tier output priceNot offered $25.00Flex tier
Cited scores
Epoch Capabilities Index— 169.1
LMArena Text— 1,475 max effort
LMArena Text · Coding— 1,548 max effort
LMArena Vision— 1,279 max effort
LMArena Text · Chinese— 1,520 max effort
Epoch AI · GPQA Diamond— 95.8% max effort
Epoch AI · FrontierMath Tiers 1–3— 93.7% max effort

Missing facts stay blank; check each source before choosing a model. How we compare →

Model details and sources

More comparisons with these models

All comparisons →

Cited scores