About
Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family. It supports text, image, and file inputs with text output, with reasoning support and a 1M-token...
Related Models
Anthropic: Claude Opus 4.8 (Fast)2026-05-27Anthropic: Claude Opus 4.7 (Fast)2026-05-12Anthropic: Claude Opus 4.72026-04-16Claude Opus 4.7 (Non-reasoning, High Effort)2026-04-16Anthropic: Claude Opus 4.6 (Fast)2026-04-07Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)2026-02-17Anthropic: Claude Sonnet 4.62026-02-17Claude Sonnet 4.6 (Non-reasoning, Low Effort)2026-02-17
Pricing
Input
$5.00
per 1M tokens
Output
$25.00
per 1M tokens
Blended
$10.00
per 1M tokens
Cheaper than 6% of models. Median price is $0.54/1M tokens.
Cost Calculator
Tokens per day1M
100K100M
Daily
$10.00
Monthly
$300.00
vs. Similar Models
OpenAI: GPT-5.5Q:-1.2
$11.25+13%
GPT-5.5 (high)Q:-2.5
$11.25+13%
Anthropic: Claude Opus 4.7Q:-4.1
$10.000%
Google: Gemini 3.1 Pro PreviewQ:-4.2
$4.50-55%
Performance
65
tokens/sec
Faster than 31% of models
12.68
seconds
Faster than 10% of models
12.68
seconds
Faster than 34% of models
Market Median
95 tok/s
31% slower
Median TTFT
1.09s
1068% slower
Throughput/Dollar
7
tok/s per $/1M
Speed Comparison
Qwen3 8B (Non-reasoning)
65 tok/s+0%
Qwen3 14B (Non-reasoning)
66 tok/s+1%
OpenAI: GPT-5.5
66 tok/s+1%
Context Window
1.0M
tokens
Larger than 82% of models
Max Output
128K
tokens
13% of context
Benchmarks
MMLU-ProNot evaluated
GPQA Diamond
92.0%
HLE
45.7%
LiveCodeBenchNot evaluated
SciCode
53.5%
TerminalBench Hard
58.3%
MATH-500Not evaluated
AIMENot evaluated
AIME 2025Not evaluated
IFBench
62.2%
Long Context Recall
67.7%
Tau2
94.4%
Market AverageTop Score