modelbenchmark.io

Claude Opus 4.5 (latest)

Anthropic · claude-opus-4-5

Compare

Flagship Claude model for deep reasoning, coding, and long-horizon agents

Specification

anthropic listing

Context
200K
Max output
64K
Released
2025-11-24
Knowledge cutoff
2025-05
Retires
2026-11-24
Open weights
no
Input
text, image, pdf
Output
text

Price

US dollars per million tokens · anthropic list price

Input
$5.00
Output
$25.00
Cache read
$0.50
Cache write
$6.25

Batch and priority prices

US dollars per million tokens · per listing

PriceListing$/M
Batch inputanthropic$2.50
Batch outputanthropic$12.50
Cache write, 1 hourbedrock_converse$10.00
Cache write, 1 hourbedrock_converse · us$11.00

Batch input is 50% below the interactive input price at the same listing.

Quality

12 benchmarks · 2 sources

Composite
46th
percentile of 344 scored models
Rank
187 / 344
±0.06 sd
Evidence
12 × 2
3 or more benchmarks, from 2 or more sources
Effort range
4.4
points between effort settings
reasoning49th4/4 bench
math46th5/7 bench
knowledge54th1/1 bench
general30th1/1 bench

By reasoning effort

every setting placed on the same scale as the leaderboard

SettingCompositeBenchmarks behind it
16k75th4
48k51st1
32k50th8
high+64k43rd1
59k27th1
128k27th1
64k24th1

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
GPQA diamond86.0 ±2.132kEpoch AI2025-11-24
↳85.5 ±2.216kEpoch AI2025-11-25
Settings differ by 0.5 points on this benchmark.
OTIS Mock AIME 2024-202586.1 ±4.332kEpoch AI2025-11-24
↳81.7 ±5.016kEpoch AI2025-11-24
Settings differ by 4.4 points on this benchmark.
SimpleQA Verified45.7 ±1.632kEpoch AI2026-08-27
FrontierMath-v130.2 ±16.316k · 2 runsEpoch AI2025-11-25
↳20.7 ±2.432kEpoch AI2025-11-25
Settings differ by 9.5 points on this benchmark.
Mystery Game Puzzles22.0 ±4.248kEpoch AI2026-07-25
LiveBench72.9high+64k · 23 runsLiveBench2026-06-25
FrontierMath-Tier-4-2025-07-014.2 ±2.932kEpoch AI2025-11-25
↳2.1 ±2.116kEpoch AI2025-11-25
Settings differ by 2.1 points on this benchmark.
FrontierMath-Tiers-1-3-v234.4 ±2.832kEpoch AI2026-06-11
EBR-bench14.3128kEpoch AI2026-06-25
Furniture Assembly28.3 ±5.964kEpoch AI2026-09-10
Chess Puzzles14.0 ±3.559kEpoch AI2026-10-13
↳5.0 ±2.232kEpoch AI2026-10-13
Settings differ by 9.0 points on this benchmark.
FrontierMath-Tier-4-v24.9 ±3.432kEpoch AI2026-06-11

The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.

Available from 51 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
anthropicvendor$5.00$25.00$0.50$6.25200K64K—
qihang-ai$0.71$3.57——200K32K—
poe$4.2929$21.4646$0.4293$5.3662197K64K—
jiekou$4.75$23.75$0.475$5.9375200K66K—
302ai$5.00$25.00——200K64K—
abacus$5.00$25.00——200K64K—
aihubmix$5.00$25.00$0.50$6.25200K64K—
amazon-bedrock$5.00$25.00$0.50$6.25200K64K—
amazon-bedrock · global$5.00$25.00$0.50$6.25200K64K—
azure$5.00$25.00$0.50$6.25200K64K—
azure_ai$5.00$25.00$0.50$6.25200K64K2026-11-24
azure-cognitive-services$5.00$25.00$0.50$6.25200K64K—
bedrock$5.00$25.00$0.50$6.25200K64K—
bedrock_converse$5.00$25.00$0.50$6.25200K64K—
bedrock_converse · global$5.00$25.00$0.50$6.25200K64K—
cloudflare-ai-gateway$5.00$25.00$0.50$6.25200K64K—
databricks$5.00$25.00$0.50$6.25200K64K—
edenai$5.00$25.00$0.50$6.25200K64K—
gmi · gmi$5.00$25.00——410K32K—
google-vertex$5.00$25.00$0.50$6.25200K64K—
google-vertex-anthropic$5.00$25.00$0.50$6.25200K64K—
impossibl$5.00$25.00$0.50$6.25200K64K—
kilo$5.00$25.00$0.50$6.25200K64K—
llmgateway$5.00$25.00$0.50$6.25200K64K—
llmgateway-providers$5.00$25.00$0.50$6.25200K32K—
llmgateway-providers · aws-bedrock$5.00$25.00$0.50$6.25200K32K—
llmgateway-providers · vertex-anthropic$5.00$25.00$0.50$6.25200K32K—
merge-gateway$5.00$25.00$0.50$6.25200K64K—
nano-gpt$5.00$25.00$0.50—200K64K—
nano-gpt · thinking$5.00$25.00$0.50—200K64K—
neon$5.00$25.00$0.50$6.25200K64K—
ofox$5.00$25.00$0.50$6.25200K64K—
opencode$5.00$25.00$0.50$6.25200K64K—
openrouter$5.00$25.00$0.50$6.25200K64K—
opper$5.00$25.00$0.50$6.25200K64K—
orcarouter$5.00$25.00$0.50$6.25200K64K—
perplexity$5.00$25.00$0.50————
perplexity-agent$5.00$25.00$0.50$6.25200K64K—
requesty$5.00$25.00$0.50$6.25200K64K—
snowflake-cortex$5.00$25.00$0.50$6.25200K64K—
tempr$5.00$25.00$0.50$6.25200K64K—
venice$5.00$25.00$0.50$6.25198K33K—
vercel$5.00$25.00$0.50$6.25200K64K—
vercel_ai_gateway · vercel-ai-gateway$5.00$25.00$0.50$6.25200K64K—
vertex_ai-anthropic_models$5.00$25.00$0.50$6.25200K64K2026-11-24
zenmux$5.00$25.00$0.50$6.25200K32K—
aimlapi$5.29$26.45$0.529$6.6125200K64K—
amazon-bedrock · eu$5.50$27.50$0.55$6.875200K64K—
amazon-bedrock · us$5.50$27.50$0.55$6.875200K64K—
bedrock_converse · eu$5.50$27.50$0.55$6.875200K64K—
bedrock_converse · us$5.50$27.50$0.55$6.875200K64K—

Price history

append-only observations · a listing writes a row only when its price moves

DateHostField$/M
2026-09-02amazon-bedrock · usCache read$0.50
2026-09-16amazon-bedrock · usCache read$0.55
2026-09-02bedrock_converseCache read$0.50
2026-09-24bedrock_converseCache read$0.55
2026-09-25bedrock_converseCache read$0.50
2026-09-02bedrock_converse · euCache read$0.50
2026-09-24bedrock_converse · euCache read$0.55
2026-09-02openrouterCache read$0.50
2026-10-10openrouterCache read$0.50
2026-09-02poeCache read$0.43
2026-10-22poeCache read$0.4293
2026-09-02veniceCache read$0.60
2026-10-15veniceCache read$0.50
2026-09-02amazon-bedrock · usCache write$6.25
2026-09-16amazon-bedrock · usCache write$6.875
2026-09-02bedrock_converseCache write$6.25
2026-09-24bedrock_converseCache write$6.875
2026-09-25bedrock_converseCache write$6.25
2026-09-02bedrock_converse · euCache write$6.25
2026-09-24bedrock_converse · euCache write$6.875
2026-09-02openrouterCache write$6.25
2026-10-10openrouterCache write$6.25
2026-09-02poeCache write$5.30
2026-10-22poeCache write$5.3662
2026-09-02veniceCache write$7.50
2026-10-15veniceCache write$6.25
2026-09-02amazon-bedrock · usInput$5.00
2026-09-16amazon-bedrock · usInput$5.50
2026-09-02bedrock_converseInput$5.00
2026-09-24bedrock_converseInput$5.50
2026-09-25bedrock_converseInput$5.00
2026-09-02bedrock_converse · euInput$5.00
2026-09-24bedrock_converse · euInput$5.50
2026-09-02jiekouInput$4.50
2026-10-22jiekouInput$4.75
2026-09-02openrouterInput$5.00
2026-10-10openrouterInput$5.00
2026-09-02poeInput$4.30
2026-10-22poeInput$4.2929
2026-09-02veniceInput$6.00
2026-10-15veniceInput$5.00
2026-09-02amazon-bedrock · usOutput$25.00
2026-09-16amazon-bedrock · usOutput$27.50
2026-09-02bedrock_converseOutput$25.00
2026-09-24bedrock_converseOutput$27.50
2026-09-25bedrock_converseOutput$25.00
2026-09-02bedrock_converse · euOutput$25.00
2026-09-24bedrock_converse · euOutput$27.50
2026-09-02jiekouOutput$22.50
2026-10-22jiekouOutput$23.75
2026-09-02openrouterOutput$25.00
2026-10-10openrouterOutput$25.00
2026-09-02poeOutput$21.00
2026-10-22poeOutput$21.4646
2026-09-02veniceOutput$30.00
2026-10-15veniceOutput$25.00
2026-10-13aimlapiCache read$0.6877
2026-10-23aimlapiCache read$0.529
2026-10-13aimlapiInput$6.877
2026-10-23aimlapiInput$5.29
2026-10-13aimlapiOutput$34.385
2026-10-23aimlapiOutput$26.45
2026-10-21aimlapiCache write$8.5962
2026-10-23aimlapiCache write$6.6125

Other listings of this model

same model, different key

More from Anthropic

most-hosted first