GPT-6 Astra

OpenAIKey gpt-6-astraAlso reported as openai/gpt-6-astra, router_client:codex/gpt-6-astra, sub2api-codex/gpt-6-astra, auto/gpt-6-astra, vm2api/gpt-6-astra, p2bc7fb/gpt-6-astra, codex/gpt-6-astra, bearlab/gpt-6-astra, owo/gpt-6-astra, linkapi/gpt-6-astra, main/gpt-6-astra, sub2api/gpt-6-astra, p00c634/gpt-6-astra, gpt/gpt-6-astra, claude-openai/gpt-6-astra, codebuddy-api/gpt-6-astra, laysoai/gpt-6-astra, router_client:unknown/gpt-6-astra, combo/gpt-6-astra, pve-codex-1/gpt-6-astra

Rank
#3
by active users
Active users
565
Down 254 from 819 active users in the previous 30 days. vs previous 30 days
Adoption
59%
of 961 accounts
Tokens
1.2T
cache included
Est. cost
$2M
at public list prices
First seen
Jul 24, 2026
in any submission

Oct 14 – Nov 12, 2026 (UTC)961 accounts100% of tokens from an identified modelData through Nov 12, 2026, 06:48 UTCHow we count

Daily active users

Distinct accounts that used this model on each UTC day.

39 active users on Nov 12, 2026

Daily active users for GPT-6 Astra, Oct 14, 2026 to Nov 12, 2026. Peak 329. Use the left and right arrow keys to read each day.4002000
Show daily values
Daily active users and tokens for GPT-6 Astra
Date (UTC)Active usersTokens
Oct 14, 202630863.6B
Oct 15, 202632956.8B
Oct 16, 202628357.4B
Oct 17, 202622951.5B
Oct 18, 202623852.3B
Oct 19, 202627152.4B
Oct 20, 202627049.5B
Oct 21, 202627739.8B
Oct 22, 202627841.2B
Oct 23, 202628255B
Oct 24, 202622351.8B
Oct 25, 202623244.6B
Oct 26, 202627251.7B
Oct 27, 202627049.4B
Oct 28, 202626651.4B
Oct 29, 202625257.8B
Oct 30, 202623240.2B
Oct 31, 202618128.1B
Nov 1, 202617225.9B
Nov 2, 202619538.3B
Nov 3, 202621248.7B
Nov 4, 202620432.9B
Nov 5, 202619733.9B
Nov 6, 202616824.8B
Nov 7, 202614615.1B
Nov 8, 202612716.8B
Nov 9, 202616320.9B
Nov 10, 202613016.7B
Nov 11, 202610111.8B
Nov 12, 2026392.9B

Token mix

How the 1.2T tokens split. Cache reads are usually most of an agent's tokens and are priced far below input.

  • Input3.0%
  • Output0.2%
  • Cache read97%
  • Cache write<0.1%
  • Reasoning<0.1%

Agents that carry GPT-6 Astra

Each agent's share of this model's tokens, and how many accounts used the model through it.

  1. Codex CLI86% of this model's tokens499active users
  2. Senpi (OmO Native)4.1% of this model's tokens42active users
  3. Oh My Pi3.2% of this model's tokens52active users
  4. Claude Code2.4% of this model's tokens16active users
  5. Pi1.7% of this model's tokens37active users
  6. OpenCode1.1% of this model's tokens28active users
  7. Hermes Agent0.7% of this model's tokens23active users
  8. Prime Agent0.5% of this model's tokens1active users
  9. Gajae Code0.2% of this model's tokens10active users
  10. Aside0.1% of this model's tokens19active users
  11. Junie<0.1% of this model's tokens1active users
  12. ZCode<0.1% of this model's tokens3active users
  13. Copilot<0.1% of this model's tokens1active users
  14. OpenClaw<0.1% of this model's tokens5active users
  15. DeepSeek Harness<0.1% of this model's tokens6active users
  16. Reasonix<0.1% of this model's tokens1active users
  17. Synthetic<0.1% of this model's tokens1active users
  18. Droid<0.1% of this model's tokens1active users
  19. Cherry Studio<0.1% of this model's tokens1active users
  20. cc-mirror/curl<0.1% of this model's tokens1active users
  21. cc-mirror/mozilla<0.1% of this model's tokens1active users
  22. 9Router<0.1% of this model's tokens1active users

Benchmarks

The best result for GPT-6 Astra in each benchmark Tokscale tracks. Each benchmark has its own unit, run conditions and source, so compare a score only with other models on the same benchmark. Rank is among the models Tokscale matched in that benchmark. Agent rows also show Tokscale usage through that agent, side by side with the score. Usage does not explain the score.

All benchmarks
  • Epoch Capabilities IndexOverall · Epoch AI
    169.0
    Rank #3 of 158 matched modelsNo conditions reportedModel released Sep 3, 2026Source: epoch.ai (opens in a new tab)
  • Terminal-Bench 4.1Agentic coding · Terminal-Bench
    60.3%
    Rank #4 of 23 matched modelsMax · CodexPublished Nov 3, 2026Source: tbench.ai (opens in a new tab)

    Tokscale usage: Codex CLI carried 86% of GPT-6 Astra tokens in the last 30 days (499 active users).

    Show 4 more results with other run conditionsHide the other results
    • Listed as GPT-6 Astra
      60.0%
      High · CodexPublished Nov 3, 2026Source: tbench.ai (opens in a new tab)
    • Listed as GPT-6 Astra
      59.7%
      XHigh · CodexPublished Nov 3, 2026Source: tbench.ai (opens in a new tab)
    • Listed as GPT-6 Astra
      55.1%
      Medium · CodexPublished Nov 3, 2026Source: tbench.ai (opens in a new tab)
    • Listed as GPT-6 Astra
      50.9%
      Low · CodexPublished Nov 3, 2026Source: tbench.ai (opens in a new tab)
  • DeepSWEAgentic coding · Datacurve, collected by Epoch AI
    74.1%
    Rank #1 of 26 matched modelsXHigh · mini-swe-agentModel released Sep 3, 2026Source: deepswe.datacurve.ai (opens in a new tab)
    Show 4 more results with other run conditionsHide the other results
    • Listed as gpt-6-astra (high)
      73.2%
      High · mini-swe-agentModel released Sep 3, 2026Source: deepswe.datacurve.ai (opens in a new tab)
    • Listed as gpt-6-astra (max)
      73.2%
      Max · mini-swe-agentModel released Sep 3, 2026Source: deepswe.datacurve.ai (opens in a new tab)
    • Listed as gpt-6-astra (medium)
      72.8%
      Medium · mini-swe-agentModel released Sep 3, 2026Source: deepswe.datacurve.ai (opens in a new tab)
    • Listed as gpt-6-astra (low)
      67.0%
      Low · mini-swe-agentModel released Sep 3, 2026Source: deepswe.datacurve.ai (opens in a new tab)
  • FrontierSWEAgentic coding · FrontierSWE, collected by Epoch AI
    65.5%
    Rank #2 of 23 matched modelsMax · proximusModel released Sep 3, 2026Source: frontierswe.com (opens in a new tab)
  • WebDev ArenaWeb development · LMArena
    1,774
    Rank #2 of 110 matched modelsMaxPublished Oct 19, 2026Source: huggingface.co (opens in a new tab)
  • Agent ArenaAgents · LMArena
    +0.15
    Rank #2 of 48 matched modelsMax · Arena Agent ModePublished Oct 19, 2026Source: huggingface.co (opens in a new tab)
  • GPQA DiamondReasoning · Epoch AI
    95.8%
    Rank #3 of 160 matched modelsMaxRun date Aug 30, 2026Source: epoch.ai (opens in a new tab)

    Near its ceiling: the top score is 96.5%, and Epoch AI estimates the reachable maximum at 97%. Differences between top models are small.

  • Humanity's Last ExamReasoning · Center for AI Safety and Scale AI, collected by Epoch AI
    54.8%
    Rank #1 of 40 matched modelsNo conditions reportedModel released Sep 3, 2026Source: lastexam.ai (opens in a new tab)
  • FrontierMath Tiers 1–3 (v2)Reasoning · Epoch AI
    93.7%
    Rank #3 of 90 matched modelsMaxRun date Aug 30, 2026Source: epoch.ai (opens in a new tab)

    Near its ceiling: the top score is 95.1%, and Epoch AI estimates the reachable maximum at 95%. Differences between top models are small.

  • SimpleQA VerifiedKnowledge · Epoch AI
    75.6%
    Rank #4 of 86 matched modelsMaxRun date Aug 30, 2026Source: epoch.ai (opens in a new tab)

Not evaluated: CursorBench. A missing result is not a score of zero.

Pricing

List prices from models.dev in US dollars per 1M tokens, retrieved Nov 12, 2026. OpenAI is the lab's own API and the reference price for the estimates below.

Price at real agent usageEstimate at list prices

What 1M tokens cost at the reference list price when they split into input, output and cache the way Tokscale users' agents used them. It is not what anyone paid.

At this model's mix
$1.44
Est. per 1M tokens
97% cache reads
At the community mix
$1.53
Est. per 1M tokens
96% cache reads
3:1 input/output blend
$20
Per 1M tokens at list prices
No cache

Cache reads were 97% of GPT-6 Astra tokens in the last 30 days and 96% of all tokens. The 3:1 blend, (3 × input + output) ÷ 4, is the usual comparison and ignores cache.

Assumptions

  • OpenAI list prices per 1M tokens: input $10, output $50, cache read $1, cache write $12.50.
  • This model's mix is its own input, output, cache and reasoning tokens in the last 30 days.
  • The community mix is the tokens of every identified model in the last 30 days.
  • This is an estimate at list prices, not what anyone paid.
  • Reasoning tokens are priced as output.
  • Long-context tiers, batch pricing and discounts are not applied.
  • Token mixes come from public Tokscale submissions. Hidden accounts are excluded.

List prices by provider

  • OpenAIReferencegpt-6-astra
    $10 input per 1M tokens$50 output per 1M tokens$1 cache read per 1M tokens$12.50 cache write per 1M tokens1.05M tokens of context
  • Amazon Bedrockopenai.gpt-6-astra
    $11 input per 1M tokens$55 output per 1M tokens$1.10 cache read per 1M tokens$13.75 cache write per 1M tokens1.05M tokens of context
  • Azuregpt-6-astra
    $10 input per 1M tokens$50 output per 1M tokens$1 cache read per 1M tokens$12.50 cache write per 1M tokens1.05M tokens of context
  • OpenRouteropenai/gpt-6-astra
    $10 input per 1M tokens$50 output per 1M tokens$1 cache read per 1M tokens$12.50 cache write per 1M tokens1.05M tokens of context
  • Vercel AI Gatewayopenai/gpt-6-astra
    $10 input per 1M tokens$50 output per 1M tokens$1 cache read per 1M tokens$12.50 cache write per 1M tokens1.05M tokens of context

Specs

From models.dev (MIT), retrieved Nov 12, 2026, as the lab or the first listed provider reports them. A dash means models.dev does not list the value.

Context window
1.05M tokens
Max output
128K tokens
Input
Text, Image, PDF
Output
Text
Knowledge cutoff
Apr 30, 2026
Release date
Sep 4, 2026
Open weights
No
Reasoning
Yes
Tool calling
Yes

Top public profiles

Public profiles that used GPT-6 Astra most in this period, by tokens. Hidden accounts are never listed.

  1. 1Vincent Koc@vincentkoc180.5Btokens$386.5Kest. cost29active days$386.5K est. cost · 29 active days
  2. 2Jeff Emanuel@Dicklesworthstone68.9Btokens$92.1Kest. cost17active days$92.1K est. cost · 17 active days
  3. 3kimchanhyung98@kimchanhyung9826.5Btokens$41Kest. cost28active days$41K est. cost · 28 active days
  4. 4Abhishek gahlot@abhishekgahlot225.3Btokens$72.3Kest. cost25active days$72.3K est. cost · 25 active days
  5. 5JsLee@JSLEEKR22.8Btokens$33.3Kest. cost26active days$33.3K est. cost · 26 active days
  6. 6Will Mitchell@StartupBros21.5Btokens$14.1Kest. cost23active days$14.1K est. cost · 23 active days
  7. 7kaptinlin@kaptinlin18.4Btokens$26.2Kest. cost23active days$26.2K est. cost · 23 active days
  8. 8Soju06@Soju0617.1Btokens$47Kest. cost28active days$47K est. cost · 28 active days
  9. 9David Strejc@david-strejc17Btokens$30.6Kest. cost9active days$30.6K est. cost · 9 active days
  10. 10dletterio1@dletterio116.3Btokens$33.5Kest. cost22active days$33.5K est. cost · 22 active days

More from OpenAI

Other OpenAI models in the same period, by active users.

  1. GPT-6.1 Sol69% adoption665active users
  2. GPT-6 Luna45% adoption434active users
  3. Codex Auto Review30% adoption289active users
  4. GPT-6.1 Terra28% adoption271active users
  5. GPT-6 Sol26% adoption247active users
  6. GPT-5.6 Sol23% adoption219active users
All models

Sources and attribution

Each score keeps its own source, unit and run conditions. Tokscale does not combine benchmarks into one score, and a missing score means the model was not evaluated, not that it scored zero.

  • Epoch AIEpoch Capabilities Index, Epoch's own runs and the external results it collects.CC BY 4.0 · Retrieved Nov 12, 2026 · Data through Nov 9, 2026Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from https://epoch.ai/benchmarks.
  • LMArenaWebDev Arena and Agent Arena leaderboards.CC BY 4.0 · Retrieved Oct 21, 2026 · Data through Oct 19, 2026LMArena leaderboard dataset (lmarena-ai/leaderboard-dataset), Hugging Face.
  • Terminal-BenchTerminal-Bench 4.1 official leaderboard rows (results only, no task content).Terms not stated · Retrieved Nov 12, 2026 · Data through Nov 11, 2026Terminal-Bench 4.1 leaderboard, tbench.ai. Every score links back to the leaderboard.
  • models.devSpecs and list prices per provider.MIT · Retrieved Nov 12, 2026models.dev, an open catalog of AI models (github.com/anomalyco/models.dev).
  • TokscaleActive users, tokens and token mix from public Tokscale submissions. Hidden accounts are excluded.How we count