Request an inviteSign in
Anthropic · closed

Opus 5.5.

Opus 5.5 by Anthropic: P1 on the Nov 3, 2026 board with a score of 58, the leader. List price $4 in and $20 out per million tokens. Test Lap: Track.

PositionP1on the latest board
Score58Artificial Analysis Intelligence Index
Price$4 / $20per 1M tokens in / out · $8 blended
Test lapTracka verdict from real work
score over time
24 daysunchanged
blended price over time
24 daysunchanged
history · Opus 5.5newest first
DayPosScore$ in / outOpenRouterVercelSpendTest lap
Nov 3, 2026P158$4 / $203.3%5.4%20%Track
Nov 2, 2026P158$4 / $203.4%5.3%19%Track
Nov 1, 2026P158$4 / $203.5%5.4%19%Track
Oct 31, 2026P158$4 / $203.6%5.5%19%Track
Oct 30, 2026P158$4 / $203.8%5.4%19%Track
Oct 29, 2026P158$4 / $204.1%5.6%19%Track
Oct 28, 2026P158$4 / $204.2%5.5%19%Track
Oct 27, 2026P158$4 / $204.3%5.5%19%Track
Oct 26, 2026P158$4 / $204.3%5.3%18%Track
Oct 25, 2026P158$4 / $204.4%5.3%17%Track
Oct 24, 2026P158$4 / $204.5%5.3%17%Track
Oct 23, 2026P158$4 / $204.5%5.4%17%Track
Oct 22, 2026P158$4 / $204.3%5.5%16%Track
Oct 21, 2026P158$4 / $204.3%5.8%16%Track
Oct 20, 2026P158$4 / $204.2%6.3%16%–
Oct 19, 2026P158$4 / $204.0%6.6%16%–
Oct 18, 2026P158$4 / $203.8%6.8%16%–
Oct 17, 2026P158$4 / $203.6%6.7%15%–
Oct 16, 2026P158$4 / $203.4%6.9%16%–
Oct 15, 2026P158$4 / $203.3%6.8%16%–
Oct 14, 2026P158$4 / $203.2%6.6%16%–
Oct 13, 2026P158$4 / $203.2%6.2%16%–
Oct 12, 2026P158$4 / $203.1%6.1%16%–
Oct 11, 2026P158$4 / $20–5.9%16%–
the spec sheet
Context window
1M tokens
Max output
128K tokens
Released
Sep 22, 2026
Inputs
text, image, pdf
Outputs
text
Cached input
$0.20 per 1M tokens
Weights
closed
questions people ask
How much does Opus 5.5 cost?

$4 per million input tokens and $20 per million output tokens at list price, $8 blended at three input tokens to one output. Cached input is $0.20 per million.

How does Opus 5.5 score?

58 on the Artificial Analysis Intelligence Index: P1 of 14 on the Nov 3, 2026 board, the leader.

What is the context window of Opus 5.5?

1M tokens, with up to 128K tokens of output.

How widely is Opus 5.5 used?

Over the trailing seven days: 3.3% of tokens on OpenRouter; 5.4% of tokens on Vercel AI Gateway; 20% of spend on Vercel AI Gateway.

Is Opus 5.5 good for real work?

Its Test Lap verdict is Track: we would send real work to it today.

the rest of the grid

Score is the Artificial Analysis Intelligence Index. Price is list, per million tokens. Usage is the share of tokens over the trailing seven days on OpenRouter and on the Vercel AI Gateway usage data, both licensed under CC BY 4.0; a dash means too small to call. Specs come from Vercel AI Gateway's model list and the lab's own documentation. Timing by sleds, which routes work across every lab and favours none.