OpenAI language model

GPT-6 Astra

OpenAI's top GPT-6 model, which OpenAI calls its most capable for the most demanding work. it takes text and images and costs five times GPT-6.1 Sol's per-token rate.

01

price

$10/M

input tokens

$50/M

output tokens

$1/M

cached input

per 1M tokens. octomind bills the listed provider price, with no markup, on every plan, including the free one (pay as you go from credits).

  • requests over 272K input tokens bill in full at 2x input and cache rates and 1.5x output; the listed price is the ≤272K rate.
  • cache writes bill at $12.50 per 1M tokens (30-minute tier); cache hits at $1.00.

02

what real work costs

workloadcost
one chat reply 3,000 tokens in, 500 out $0.06
summarise a 100-page report 50,000 tokens in, 1,500 out, one request $0.58
1,000 classification calls 500 tokens in, 20 out, per call $6.00
coding-agent session 2M tokens in across ~60 turns, 60,000 out, no cache $23.00
same session, cached 85% of input read from the prompt cache, the rest written to it $8.45

at the listed rate. token counts are typical sizes, not measurements.

03

specs

provider
OpenAI
context window
1M tokens — ≈ 1,500 pages of prose or 100,000 lines of code
max output
128k tokens
input
text, images
output
text
structured output
yes — JSON schema
reasoning mode
yes
prompt caching
yes — cache hits at $1/M
weights
closed
released
September 2026
knowledge cutoff
April 2026
status
live
model id
gpt-6-astra

integration notes

  • reasoning cannot be turned off: the none effort is not supported, so the lowest effort is low.
  • tool calling requires the Responses API; Chat Completions only accepts requests without tools.
  • remove temperature, top_p and top_logprobs from requests; Chat Completions requests must also drop logprobs.

04

benchmarks

as published by the maker; not measured by octomind.

benchmarkscore
Terminal-Bench 4.0source57.9%
GPQA Diamondsource96.0%
Humanity's Last Exam (with tools)source57.2%
BrowseCompsource91.5%
OSWorld 2.0 (v2026.08.08, offline set, partial score)source72.6%

05

when to pick it

pick it if

  • you want OpenAI's top tier for complex reasoning, coding or computer use
  • you resend long prompts; cache hits bill at 10% of input
  • you need replies of up to 128K output tokens

06

where it sits on price

blended price $20/M at 3 input tokens per output token: 43 live language models cost less, 0 cost more.

  1. Kimi K3 $6/M
  2. Claude Opus 5.5 $8/M
  3. Muse Wonder 1.0 $8/M
  4. Claude Fable 5.2 $16/M
  5. GPT-6 Astra $20/M

07

use GPT-6 Astra in octomind

one octohub login reaches every model in the catalog. pick this one per run, per session, or as your default.

run it
octomind login
octomind run -m octohub:gpt-6-astra
switch mid-session
/model octohub:gpt-6-astra
make it the default — config.toml
[model]
name = "octohub:gpt-6-astra"

08

compare GPT-6 Astra

all model comparisons →

09

similar models

nearest live models by listed input price.

view all models →

10

GPT-6 Astra faq

how much does GPT-6 Astra cost?

GPT-6 Astra lists at $10/M input, $50/M output, $1/M cached input per 1M tokens. a typical coding-agent session (2M tokens in, 60,000 out) comes to $23.00, or $8.45 when 85% of the input is read from the prompt cache. octomind bills the listed provider price, with no markup, on every plan, including the free one (pay as you go from credits).

what is the context window of GPT-6 Astra?

1M tokens, ≈ 1,500 pages of prose or 100,000 lines of code. a single reply can run to 128k tokens.

does GPT-6 Astra accept images?

yes. it takes text and images as input and returns text.

is GPT-6 Astra open source?

no. GPT-6 Astra is a closed-weights model, available only through OpenAI's api and hosted gateways such as octohub.

when was GPT-6 Astra released?

September 2026, by OpenAI.

is GPT-6 Astra free?

no. GPT-6 Astra bills per token at $10/M input, $50/M output, $1/M cached input per 1M. octomind's free plan reaches it pay as you go from credits, so you pay only for the tokens you use, at the provider price. credits carry a 3% top-up fee.

how do i use GPT-6 Astra in octomind?

run octomind login once, then octomind run -m octohub:gpt-6-astra. inside a running session, /model octohub:gpt-6-astra switches to it without losing context.

use GPT-6 Astra in octomind

choose a model when you start, then switch mid-session when the work changes.