modelbenchmark.io

Claude Opus 4.8

Anthropic · claude-opus-4-8

Compare

Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents

Specification

anthropic listing

Context
1M
Max output
128K
Released
2026-05-28
Knowledge cutoff
2026-01
Retires
2027-05-28
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
Cache write, 1 hourbedrock_converse$10.00
Cache write, 1 hourbedrock_converse · us$11.00
Cache write, 1 hourbedrock$12.00

Quality

13 benchmarks · 2 sources

Composite
89th
percentile of 334 scored models
Rank
38 / 334
±0.06 sd
Evidence
13 × 2
3 or more benchmarks, from 2 or more sources
Effort range
3.8
points between effort settings
reasoning88th4/4 bench
math92nd6/7 bench
knowledge79th1/1 bench
general59th1/1 bench

By reasoning effort

every setting placed on the same scale as the leaderboard

SettingCompositeBenchmarks behind it
low93rd3
max91st11
xhigh87th1
high63rd1

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
FrontierMath-Tiers-1-3-v280.0 ±2.4maxEpoch AI2026-06-10
FrontierMath-Tier-4-2025-07-0131.2 ±6.8maxEpoch AI2026-06-08
OTIS Mock AIME 2024-202598.3 ±1.4maxEpoch AI2026-06-07
97.8 ±2.2lowEpoch AI2026-08-06
Settings differ by 0.5 points on this benchmark.
GPQA diamond91.0 ±1.9maxEpoch AI2026-06-07
88.4 ±2.3lowEpoch AI2026-08-06
Settings differ by 2.6 points on this benchmark.
FrontierMath-v147.2 ±2.9maxEpoch AI2026-06-08
Chess Puzzles34.0 ±4.8maxEpoch AI2026-05-29
29.0 ±4.6lowEpoch AI2026-08-06
Settings differ by 5.0 points on this benchmark.
Mystery Game Puzzles36.0 ±4.8maxEpoch AI2026-07-25
31.0 ±4.6xhighEpoch AI2026-07-26
Settings differ by 5.0 points on this benchmark.
FrontierMath-Tier-4-v256.1 ±7.8maxEpoch AI2026-06-10
SimpleQA Verified53.0 ±1.6maxEpoch AI2026-08-27
LiveBench77.1max · 23 runsLiveBench2026-06-25
EBR-bench28.6 ±1.4maxEpoch AI2026-08-07
OEIS Open Lite39.0 ±4.9highEpoch AI2026-08-11
OEIS Open29.9 ±2.1highEpoch AI2026-08-11

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

Available from 62 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
anthropicvendor$5.00$25.00$0.50$6.251M128K2027-05-28
kenarifreefree1M128K
unorouter$0.425$2.1251M128K
xpersona$1.50$9.25$0.15200K128K
kilo · stealth$4.00$20.00$0.40$5.001M128K
poe$4.2929$21.46461M128K
302ai$5.00$25.001M128K
abacus$5.00$25.001M128K
aihubmix$5.00$25.00$0.50$6.25200K32K
amazon-bedrock$5.00$25.00$0.50$6.251M128K
amazon-bedrock · au$5.00$25.00$0.50$6.251M128K
amazon-bedrock · global$5.00$25.00$0.50$6.251M128K
amazon-bedrock · jp$5.00$25.00$0.50$6.251M128K
amazon-bedrock · us$5.00$25.00$0.50$6.251M128K
azure$5.00$25.00$0.50$6.251M128K
azure_ai$5.00$25.00$0.50$6.251M128K2027-09-01
azure-cognitive-services$5.00$25.00$0.50$6.251M128K
bedrock_converse$5.00$25.00$0.50$6.251M128K
bedrock_converse · global$5.00$25.00$0.50$6.251M128K
cloudflare-ai-gateway$5.00$25.00$0.50$6.251M128K
crossmodel$5.00$25.00$0.50$6.251M128K
daoxe$5.00$25.00$0.50$6.251M128K
databricks$5.00$25.00$0.50$6.251M128K
deepinfra$5.00$25.001M
edenai$5.00$25.00$0.50$6.251M128K
fastrouter$5.00$25.001M128K
freemodel$5.00$25.00$0.50$6.251M128K
github-copilot$5.00$25.00$0.50$6.25200K64K
gmicloud$5.00$25.00$0.501M128K
impossibl$5.00$25.00$0.50$6.251M128K
kilo$5.00$25.00$0.50$6.251M128K
llmgateway$5.00$25.00$0.50$6.251M128K
llmgateway-providers$5.00$25.00$0.50$6.251M128K
llmgateway-providers · aws-bedrock$5.00$25.00$0.50$6.251M128K
llmgateway-providers · azure-anthropic$5.00$25.00$0.50$6.251M128K
merge-gateway$5.00$25.00$0.50$6.251M128K
modelis$5.00$25.001M128K
nano-gpt$5.00$25.00$0.501M128K
nano-gpt · thinking$5.00$25.00$0.501M128K
neon$5.00$25.00$0.50$6.251M128K
ofox$5.00$25.00$0.50$6.251M128K
opencode$5.00$25.00$0.50$6.251M128K
openrouter$5.00$25.00$0.50$6.251M128K
opper$5.00$25.00$0.50$6.251M128K
orcarouter$5.00$25.00$0.50$6.251M128K
perplexity$5.00$25.00$0.50
pioneer$5.00$25.00$0.50$6.251M128K
requesty$5.00$25.00$0.50$6.251M128K
routing-run$5.00$25.001M32K
vercel$5.00$25.00$0.50$6.251M128K
vertex_ai-anthropic_models$5.00$25.00$0.50$6.251M128K2027-05-28
zenmux$5.00$25.00$0.50$6.251M128K
amazon-bedrock · eu$5.50$27.50$0.55$6.8751M128K
bedrock_converse · au$5.50$27.50$0.55$6.8751M128K
bedrock_converse · eu$5.50$27.50$0.55$6.8751M128K
bedrock_converse · jp$5.50$27.50$0.55$6.8751M128K
bedrock_converse · us$5.50$27.50$0.55$6.8751M128K
bedrock$6.00$30.00$0.60$7.501M128K
venice$6.00$30.00$0.60$7.501M128K
agentrouter1M128K
model-oracle-ai1M128K
snowflake-cortex1M128K

Price history

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

DateHostField$/M
2026-09-02openrouterCache read$0.50
2026-09-06openrouterCache read$0.50
2026-09-02openrouterCache write$6.25
2026-09-06openrouterCache write$6.25
2026-09-02openrouterInput$5.00
2026-09-06openrouterInput$5.00
2026-09-02openrouterOutput$25.00
2026-09-06openrouterOutput$25.00

Other listings of this model

same model, different key

More from Anthropic

most-hosted first