LLM

Claude Opus 4.6

by Anthropic

Prices last updated November 17, 2026

Anthropic's most capable model with 1M token context and advanced reasoning.

Context
1M
Providers
12
Cheapest Input
$1.43
Vision Function Calling JSON Mode Streaming Chat / InstructCodeVision

Price Comparison

Compare pricing across 12 providers serving Claude Opus 4.6

Provider Input $/1M Output $/1M Context Speed TTFT Status
KIE AI
KIE AI
$1.43 ● $7.15 — — — Live
PoYo
PoYo
$2.70 $12.00 — — — Live
Muapi
Muapi
$3.00 $15.00 — — — Live
Replicate
Replicate
$5.00 $25.00 — — — Live
OpenRouter
OpenRouter
$5.00 $25.00 — — — Live
Anthropic
Anthropic
$5.00 $25.00 — — — Live
WaveSpeed AI
WaveSpeed AI
$5.00 $25.00 — — — Live
Amazon Bedrock
Amazon Bedrock
$5.00 $25.00 1M — — Live
Venice AI
Venice AI
$5.00 $25.00 — — — Live
Atlas Cloud
Atlas Cloud
$5.00 $25.00 1M — — Live
Infron
Infron
$5.00 $25.00 1M — — Live
AIMLAPI
AIMLAPI
$6.50 $32.50 — — — Live
● Cheapest ● Fastest Prices in USD per 1M tokens

Benchmarks

Independent results for Claude Opus 4.6, as each source publishes them. Where a source tested several versions (effort levels, thinking modes, dated snapshots), the best one is shown and named under "As tested". Scores from different sources use different scales and aren't comparable with each other.

BenchmarkScoreRankAs tested
Overall
One score combining a model's results across many benchmarks, on a common scale.
156.3±1.8
#40 of 258
Claude Opus 4.6
Reasoning
GPQA DiamondEpoch AI
Graduate-level multiple-choice questions in biology, physics and chemistry, written to be hard to look up.
90.5%±3.3
#47 of 232
claude-opus-4-6_32K
Feb 2026
Math
Unpublished research-level maths problems with automatically checkable answers.
66.0%±5.5
#40 of 95
claude-opus-4-6_max
Jun 2026
FrontierMath's hardest tier: problems that take specialist mathematicians days.
26.8%±13.7
#44 of 77
claude-opus-4-6_max
Jun 2026
Competition maths problems from the OTIS mock AIME exams (2024-2025).
94.4%±5.5
#55 of 215
claude-opus-4-6_64K
Feb 2026
Coding
Real GitHub issues from Python projects; the model must write a patch that passes the project's tests. Epoch's own runs.
78.7%±3.7
#4 of 32
claude-opus-4-6
Feb 2026
Text Arena votes on coding prompts only.
1,547±5
21,650 votes
#9 of 421
Claude Opus 4.6 (High)
Nov 2026
Human votes on web apps the models build from a prompt.
1,546±6
19,013 votes
#58 of 161
Claude Opus 4.6 (High)
Nov 2026
Knowledge
Short factual questions answered without search, measuring what the model knows and whether it makes things up.
47.0%±3.1
#41 of 97
claude-opus-4-6_max
Aug 2026
Chat
Text ArenaLMArena
Human votes on chat answers across all kinds of prompts (style-controlled).
1,503±3
84,188 votes
#8 of 426
Claude Opus 4.6 (High)
Nov 2026
Vision
Human votes on answers to prompts that include images (style-controlled).
1,298±6
22,067 votes
#4 of 168
Claude Opus 4.6 (High)
Nov 2026

Epoch AI: Epoch runs each benchmark itself on the models it evaluates, with the same harness for every model. The Epoch Capabilities Index combines many benchmarks into one score. Data: Epoch AI, 'Capabilities & benchmarking', epoch.ai/benchmarks (CC BY 4.0).

LMArena: People compare two anonymous models side by side and vote for the better answer; ratings come from those votes (Bradley-Terry, style-controlled for text and vision). Data: LMArena (arena.ai) leaderboard dataset (CC BY 4.0).

Rank is among all models the source ranks, including ones we don't list.

Claude Opus 4.6 guides

Subscribe to our newsletter

News and updates, 1–2 emails per month. No spam.