Anthropic

Claude 4.5 Opus

Type
language model
Context
200K tokens
Max output
32K
Released
1 November 2025
Knowledge cutoff
March 2025
API string
claude-opus-4-5-20251101

Prices by provider

Provider Input, $ per 1M Output, $ per 1M In our data since
qihang-ai $0.71 $3.57 31 Jul 2026
Jiekou.AI $4.5 $22.5 31 Jul 2026
NanoGPT $4.998 $25.007 31 Jul 2026
302.AI $5 $25 31 Jul 2026
Abacus.AI $5 $25 31 Jul 2026
Amazon Bedrock anthropic.claude-opus-4-5-20251101-v1:0 $5 $25 2 Aug 2026
Anthropic $5 $25 31 Jul 2026
LLM Gateway $5 $25 31 Jul 2026
Merge Gateway $5 $25 31 Jul 2026
Amazon Bedrock eu.anthropic.claude-opus-4-5-20251101-v1:0 $5.5 $27.5 2 Aug 2026

Only the base tier and only per-token prices are shown. Batch, discounted and cached rates, as well as prices per image or per second of video, do not go into this table: they cannot stand in the same column as a price per million tokens.

A single provider appears several times if it sells this model under different identifiers and at different prices — for example, at different weight precision. The identifier is shown next to the name. Identical offers that arrived from two sources under different spellings are merged into one row.

The date in the last column is the day this price first entered our collection. The price may well be older: before that day we simply were not recording it. It has not changed since — otherwise a new row with a new date would stand in its place.

Measurement results

Task set Result Run conditions Measured by
Arena Score, programming 1,498.02 1,493–1,503
Arena Score, expert questions 1,486.32 1,477–1,495
Arena Score, long queries 1,478.41 1,473–1,483
Arena Score in French 1,477.28 1,461–1,493
Arena Score, hard prompts 1,474.98 1,471–1,479
Arena Score, instruction following 1,471.36 1,466–1,476
Arena Score, web development 1,466.99 1,460–1,474
Arena Score, multi-turn dialogue 1,464.55 1,458–1,471
Arena Score in English 1,460.01 1,456–1,464
Arena Score, mathematics 1,459.45 1,450–1,469
Arena Score in Spanish 1,453.21 1,438–1,468
Arena Score, overall 1,449.24 1,446–1,452
Arena Score in Russian 1,446.81 1,439–1,454
Arena Score, creative writing 1,440.86 1,434–1,447
Mock AIME 2024–2025 — olympiad problems 86.1 % · with a tuned harness Epoch evaluations
Cybench — cybersecurity tasks 82 % · with a tuned harness Opus 4.5 System Card self-reported
GPQA Diamond — graduate-level questions 81.4 % thinking budget: 16K · with a tuned harness Epoch evaluations
ARC-AGI — generalising to unseen patterns 80 % thinking budget: 64K · with a tuned harness ARC Prize Leaderboard
SWE-bench Verified — fixing bugs in repositories 76.65 % · with a tuned harness Epoch evaluations
GeoBench — locating a place from a photograph 75 % GeoBench leaderboard
OSWorld — working inside an operating system, first version 66.3 % · with a tuned harness Anthropic annoucement self-reported
WeirdML — unusual machine learning tasks 63.7 % thinking budget: 16K WeirdML Leaderboard
Terminal-Bench — working in the command line 63.1 % · with a tuned harness Terminal-Bench v2 Leaderboard
SimpleBench — trick questions 54.4 % SimpleBench Leaderboard
GDPval — tasks from real occupations 45.5 %
BALROG — game environments 43.5 % Balrog Leaderboard
SimpleQA Verified — factual accuracy 41.8 % thinking budget: 32K Epoch evaluations
ARC-AGI-2 37.64 % thinking budget: 64K
FrontierMath, levels 1–3 34.39 % thinking budget: 32K Epoch evaluations
GSO-Bench — code optimisation 26.5 % GSO Leaderboard
Humanity’s Last Exam — expert-level questions 21.43 %
APEX-Agents 20.7 %
Chess puzzles 7.41 % thinking budget: 32K Epoch evaluations
FrontierMath, level 4 — research-grade problems 4.88 % thinking budget: 32K Epoch evaluations
Remote Labor Index — jobs from a freelance marketplace 3.75 %