Skip to content
ModelCap

GPT-6.2 Astra

NewPreliminaryNo linked weights1.5d #21 → #22

OpenAI·#22 of 371 on the current board·cheaper than #15 GPT-5.5 Pro

GPT-6.2 Astra is a language model from OpenAI. It ranks #22 of 371 on ModelCap's current board, with a preliminary Index score of 77.9, partly measured on public benchmarks and partly modeled. Its score range of 67.4–88.4 spans positions #1–#51. It is listed at $8.00 per million input tokens and $40.00 per million output tokens across 3 providers. It reads up to 1M tokens of context, with no downloadable weights linked in the current source data.

Figures as of 19 Nov 2026, 05:16 UTC · download the public dataset · methodology · launch report

ModelCap Index · Current board
77.9#22
Partly measured on public benchmarks and partly modeled · range 67.4–88.4
Evidence as of 18 Nov 2026, 14:36 UTC
Market Gravity
45.1Increased by 1.2
Secondary market signal
Input
$8.00
per 1M tokens
Output
$40.00
per 1M tokens
Context
1M
tokens
Providers
3
5 endpoints

Providers

Uptime measured over the last 30 minutes
ProviderInputOutputContextQuantUptime
Amazon BedrockglobalCheapest$8.00$40.001M—100.0%
Azure$8.00$40.001M—100.0%
OpenAI$8.00$40.001M—100.0%
OpenAIfast$16.00$80.001M—100.0%
OpenAIultrafast$24.00$1201M—100.0%

Reasoning efforts

Best supported result can inform ModelCap

Published Max, X-High, High and other configurations for this model. These are sourced scores, not head-to-head essays.

EffortAgent scoreSource rankSampleSource
Max
GPT-6.2 Astra (Max)
15.22%
CI 12.5–18.0%
#6430K obs
5K sessions
LMArena Agent
Match 100%

Other versions (1)

ModelCap Index · Current board

Methodology
77.9
Partly measured on public benchmarks and partly modeled
Current board
#22 of 371

Preliminary. Measured on 3 public benchmarks so far, with limited support. Expect the score and position to move as more boards report. What this label means

Score range
67.4–88.4
Rank range
#1–#51
Evidence as of
18 Nov 2026, 14:36 UTC
Benchmarks used
ARC-AGI-2 Verified, ARC-AGI-3 Verified, LMArena Agent
Technical details
Public rank basis
Measured · blended (measured)
Evidence label
Measured · blended · 1 reported rows against measured corpus ladders (20%); publisher-corpus-prior over 214 held-out anchors (13%); opens above measured predecessor openai/gpt-6-astra on same-family ordering evidence (47%); 3 specialist-board observations at 4.9% support (21%); shrunk 3.8 toward the measured corpus
Index support
4.9% · weak
Identity binding
Exact catalogue product · openai/openai/gpt-6.2-astra · aggregates disclosed benchmark configurations · endpoint configuration not claimed
Index observations used
3
Qualified Index evidence point
77.9
How this position was produced · fused launch placement

No general-family board lists this model yet. Its point estimate is the precision-weighted mean of every channel that spoke for it; each channel’s share is the weight it carried. Independent measurements replace this placement as boards list the model.

Fused estimate
77.9 ± 8.2
Independent share
21%
Shrink toward corpus
-3.8
Optimism haircut
None applied
Method
precision-weighted-v1
Channels fused into this model’s launch placement
ChannelEstimateShare
Corpus ladder · not independent
1 reported rows against measured corpus ladders
86.8 ± 18.520%
Publisher corpus prior · not independent
publisher-corpus-prior over 214 held-out anchors
52.1 ± 22.913%
Family succession · not independent
opens above measured predecessor openai/gpt-6-astra on same-family ordering evidence
78.7 ± 12.047%
Specialist board measurement
3 specialist-board observations at 4.9% support
83.7 ± 18.121%
Bayesian Evidence Score · measured evidence input only

BES normalizes admitted public benchmark evidence for measured Index rows. Its legacy score and observables prior do not define the public language rank.

Observed capability
83.7
Evidence status
provisional · 13% mass
General preference
—
Coding
—
Agents & tools
82.1
Reasoning
86.3
Evidence breadth
8%
Evidence coverage
5%
Market and catalogue signals · secondary, never the language Index rank
Usage (OpenRouter popularity)
20.4
Liquidity (providers × uptime)
85.7
Open reach (HF downloads)
—
Surface (context / tools / modalities)
100.0
Freshness
99.4
  • · OpenRouter popularity #105
  • · 3 live providers
  • · Catalogue: context 1050000, reasoning, tools, modalities File/Image/Text
  • · First seen 2026-11-17

These adoption and deployment observations remain context only. They do not change this model's ModelCap Index score or rank.

BES evidence selected · measured input only

The bounded 0–100 capability score blends each source’s competitive placement with its published achievement, then combines capability families. Evidence breadth adds a modest uncertainty adjustment; price and popularity are not part of benchmark-led rank.

Deployment readiness

Metadata index
50.0
partial · 56% metadata coverage
Not capability
Artifact reproducibility
0.0
Access & legal clarity
30.3
Deployability
100.0
Evaluation provenance
69.7

Not yet published: a pinned repository revision, the architecture and model type, a parameter count, safetensors weight metadata, the base model it derives from, a declared licence, a link to the licence terms, pinned evaluation dataset versions.

A metadata completeness and deployability index—not a safety certification, quality grade, or production-readiness claim.

Market activity

Venue coverage
Market Gravity
45.1
OpenRouter weekly popularity
#105 of 345

Market Gravity uses OpenRouter’s full-catalogue popularity order. Sparse Vercel top-ten observations appear only when published and do not affect the score.

Arena preference

No matched Arena rating

This model is visible because it is available in the live catalogue, but it has no trustworthy community result yet. ModelCap does not infer a quality score from price, features, or market activity.

Market Gravity breakdown

Usage55%
20.4

OpenRouter weekly popularity across the full model catalogue.

Liquidity25%
85.7

Independent providers versus the model's open or closed cohort, weighted by uptime.

Open reach15%
50.0

Hugging Face 30-day downloads for open models; neutral for closed models.

Freshness5%
99.4

Time since first public availability, on a six-month half-life.

Specification

Model ID
openai/gpt-6.2-astra
Context window
1M tokens
Declared output cap
128K tokens

OpenRouter catalogue or provider-endpoint declaration. Limits can vary by provider and prompt length; a model-wide maximum has not been independently verified. OpenRouter metadata

Inputs
File, Image, Text
Outputs
Text
Tokenizer
GPT
Cached input tokens
$0.40 / 1M
First seen on OpenRouter
17 Nov 2026

Capabilities

  • Supported: Reasoning
  • Supported: Tool use
  • Supported: Structured output
  • Supported: Response format
  • Supported: Moderated

Compare GPT-6.2 Astra with

Pick any model

Benchmark leaderboards featuring GPT-6.2 Astra

All leaderboards
Public benchmark boards on which GPT-6.2 Astra has a published result
LeaderboardScoreSource rankConfigurationPublished
LMArena Agent leaderboardLMArena15.2%12.5%–18.0%#6 of 68Max18 Nov 2026
ARC-AGI-2 leaderboardARC Prize Foundation95.8%#1 of 299Max18 Nov 2026
ARC-AGI-3 leaderboardARC Prize Foundation70.3%#36 of 151Max18 Nov 2026
Artificial Analysis Intelligence IndexArtificial Analysis62.1#3 of 125Max17 Nov 2026

Alternatives to GPT-6.2 Astra, by the numbers

Ranked neighbours

Models within 3 places of #22 on the ModelCap Index.

Cheaper at a similar or higher Index

Index score within 3 points of, or above, this model's, with a lower listed blended price (3:1 input:output).

Other ranked OpenAI models

The same publisher's other current models with a public Index position.

GPT-6.2 Astra: common questions

How does GPT-6.2 Astra rank among AI models?

GPT-6.2 Astra holds ModelCap Index position #22 of 371 ranked language models as of 19 November 2026, with an Index score of 77.9 (interval 67.4–88.4); evidence: measured · blended. The Index combines public benchmark boards with published uncertainty; the methodology page explains the weighting.

How much does GPT-6.2 Astra cost per 1M tokens?

GPT-6.2 Astra is listed at $8.00 per 1M input tokens and $40.00 per 1M output tokens as of 19 November 2026 across 3 providers. Prices come from the OpenRouter catalogue and re-observe on every refresh. For 1,000 requests using 2,000 input and 500 output tokens each, the listed rates imply $36.00 total (2M input + 0.5M output). This excludes caching, batch discounts, prompt-length tiers, tool charges and retries; verify the chosen endpoint's rate and limits before budgeting.

What is the context window of GPT-6.2 Astra?

GPT-6.2 Astra has a published context window of 1M tokens (1,050,000). The snapshot lists a 128,000 token catalogue-declared route output cap. OpenRouter catalogue or provider-endpoint declaration. Limits can vary by provider and prompt length; a model-wide maximum has not been independently verified. The providers table lists each endpoint's context limit.

Is GPT-6.2 Astra open-weight?

No linked weights. No downloadable model weights are linked in the current source data. This does not identify every API or chat product that offers the model.

Which API providers serve GPT-6.2 Astra?

3 providers serve GPT-6.2 Astra through OpenRouter as of 19 November 2026: OpenAI, Azure and Amazon Bedrock. Each provider's price, context limit, quantization and measured uptime are in the providers table above.

Which benchmarks has GPT-6.2 Astra been evaluated on?

GPT-6.2 Astra has published results on 4 public boards tracked by ModelCap as of 19 November 2026: LMArena Agent 15.2% (#6 of 68), ARC-AGI-2 95.8% (#1 of 299), ARC-AGI-3 70.3% (#36 of 151) and AA Intelligence Index 62.1 (#3 of 125). Each board page ranks every tracked model on that benchmark. The ModelCap Index scores the results its methodology admits, with published uncertainty, and shows reference boards without scoring them.

How does GPT-6.2 Astra compare with Claude Sonnet 5.5?

GPT-6.2 Astra ranks #22 (Index 77.9) and Claude Sonnet 5.5 ranks #24 (Index 77.3) on the ModelCap Index as of 19 November 2026. The head-to-head page lines up their benchmarks, listed prices, context windows and provider counts side by side.

When was GPT-6.2 Astra released?

GPT-6.2 Astra first appeared in the catalogue on 17 Nov 2026. It is the current version in its family. Rank and price movements since then are recorded on the site's changes feed.