GPT-6 Astra

OpenAIProprietaryReleased Sep 3, 2026

On the arena.ai Agents leaderboard, GPT-6 Astra has a Net Improvement of 14.24% (Max, rank #5, 95% interval 12.21%–16.27%). GPT-6 Astra ranks #41 on the LMArena Text leaderboard (style control) with an Arena score of 1474.1 (±5.6, 18,315 votes). Artificial Analysis Intelligence Index: 56.9. API price: $10.00 input / $50.00 output per 1M tokens. Context: 1.05M. Nearest rivals: GLM-5.3 Flash (1474.2) above and Qwen3.7 Max (1473.8) below.

Quick answer

Agents leaderboard
#5 of 37 · 14.24% success
Text (LMArena)
#41 · 1474.1
17NAS overall
#5 of 36
API price (in / out)
$10.00 / $50.00 per 1M tokens
Data updated
Nov 10, 2026 · synced daily

Sources: arena.ai, LMArena, Artificial Analysis, OpenRouter

Key metrics

Agent Net Improvement14.24%Max15,520 sessions · #5
Arena score1474.1±5.6LMArena #41 · 18,315 votes
Intelligence56.9Variant: max
Coding57.8Artificial Analysis
Writing (creative)1420.7±10.4LMArena creative writing #69 / 419
List price in / out$10.00 / $50.00USD per 1M tokens
Lowest route$10.00 / $50.00Cheapest OpenRouter route
Context1.05Mtokens
Output speed51tok/sAA measured median
TTFT / first answer222 / 222sReasoning model: includes thinking

Rank history

LMArena rank, last 26 days
#29#36#42Oct 16Nov 10

#30 on Oct 16, #41 now; best #30 in this period.

Reasoning effort comparison

Artificial Analysis measurements per reasoning effort of the same model
EffortIntelligenceCodingtok/sTTFT (s)First answer (s)
maxUsed in table 56.957.851222222
xhigh 55.357.6519191
high 54.854.7432727
medium 51.651.8463.93.9
low 43.348.0451.51.5

Models ranked nearby

Sorted by Arena score
#ModelArenaIntel.Output $tok/s
30GLM-5.21475.1–$4.40–
31DeepSeek V4.1 Flash1475.0–$1.20–
32GLM-5.3 Flash1474.227.8$0.5062
33GPT-6 AstraThis page1474.156.9$50.0051
34Qwen3.7 Max1473.8–$7.50–
35Gemini 3 Flash1472.9–$3.00–
36Claude Sonnet 5.51472.554.5$10.00133

More from OpenAI

#ModelArenaIntel.Output $tok/s
20GPT-5.6 Sol1484.9–$20.00–
24GPT-6.1 Sol1482.154.6$10.0061
26GPT-5.51481.3–$30.00–
28GPT-5.2 Chat1475.7–$14.00–
29GPT-5.41475.2–$15.00–
39GPT-5.6 Terra1467.932.3$12.00–

Pricing

  • Official API list price: $10.00 input and $50.00 output per 1M tokens (USD).
  • At a typical 3:1 input-to-output ratio, the blended cost is about $20.00 per 1M tokens.
  • Among 79 models with an output price, it is the #76 cheapest.
  • The cheapest OpenRouter route charges $10.00 input / $50.00 output.

Sources & methodology

On the arena.ai Agents leaderboard, GPT-6 Astra has a Net Improvement of 14.24% (Max, rank #5, 95% interval 12.21%–16.27%).

  • arena.ai Agents — agent Net Improvement, 95% interval, sessions and rank (via the public Hugging Face dataset, CC BY 4.0).
  • LMArena — Arena score (Text leaderboard with style control, the same view arena.ai shows by default), 95% confidence interval, votes and rank.
  • Artificial Analysis — Intelligence / Coding Index, output speed and latency (medians).
  • OpenRouter — Route prices, context length and listing date.

Arena scores use the Bradley-Terry model (not Elo); ± is the half-width of the 95% confidence interval. Latency figures are Artificial Analysis medians; for reasoning models, time to first token and to first answer include the thinking phase. Data syncs daily at 14:20 (UTC+8).

Limits: Arena scores come from anonymous users voting on chats, so they reflect conversational preference rather than agent or coding benchmarks; with few votes the interval is wide and ranks move. Prices and speeds change as providers update them.

FAQ

How does GPT-6 Astra do on the Agents leaderboard?

On the arena.ai Agents leaderboard, GPT-6 Astra has a Net Improvement of 14.24% (Max, rank #5, 95% interval 12.21%–16.27%).

What rank is GPT-6 Astra right now?

GPT-6 Astra is #33 on our leaderboard of 70 scored models (Arena score 1474.1); its raw LMArena rank is #41.

What does GPT-6 Astra cost via API?

Input $10.00 / output $50.00 per 1M tokens (official API list price; the OpenRouter quote when no official price is available). At a typical 3:1 input-to-output ratio, the blended cost is about $20.00 per 1M tokens.

How fast is GPT-6 Astra?

Artificial Analysis measures about 51 output tokens per second, 222 s to the first token and 222 s to the first answer token; it is a reasoning model, so latency includes thinking time.

How close is GPT-6 Astra to the models ranked next to it?

It is 0.1 points behind GLM-5.3 Flash (1474.2) and 0.3 points ahead of Qwen3.7 Max (1473.8); gaps smaller than the confidence interval (±5.6) are not significant.

Is GPT-6 Astra open source? Can it be self-hosted?

GPT-6 Astra is a closed model, available only via official APIs or hosted services.

What is the context length of GPT-6 Astra?

1.05M (1,050,000 tokens, as listed on OpenRouter).

GPT-6 Astra vs GLM-5.3 Flash: which is better?

On the LMArena Text leaderboard GPT-6 Astra scores 1474.1 and GLM-5.3 Flash scores 1474.2 (a 0.1-point gap, within the confidence interval, so not significant). Agent task success: GPT-6 Astra 14.24%, GLM-5.3 Flash -2.80%. On price, GPT-6 Astra costs $50.00 and GLM-5.3 Flash $0.50 per 1M output tokens.

Is there a cheaper model close to GPT-6 Astra?

Yes: Kimi K3.1 Pro scores 1488.0 on the Text leaderboard (GPT-6 Astra: 1474.1) and costs $10.00 per 1M output tokens, about 80% less than GPT-6 Astra ($50.00).

Where does the GPT-6 Astra data come from and how often is it updated?

Agent success rates come from the arena.ai Agents leaderboard (CC BY 4.0), Text scores from LMArena, intelligence and speed from Artificial Analysis, and prices and context from official list prices and OpenRouter. 17NAS syncs daily at 06:20 UTC; this page was last updated Nov 10, 2026.

Related leaderboards