Skip to content
OSMEXO
Sign inGet an API key
☰
Sign inGet an API key

Claude Opus 4.8

anthropic/claude-opus-4.8
  • Anthropic
  • Family · Claude Opus 4
  • Released 28 May 2026
  • Text · Image · File → Text
  • Reasoning · adjustable
  • Tools · JSON · Vision
Not on OSMEXO yet
Request this model
Spec & list price
Context1M
Max output128K
Input / 1M$5.00
Output / 1M$25.005× input
Scores · standing
Intelligence41.8#29/489
Coding74.3#15/196
Usage
Tokens / day9.1B▼ 22% 7d
Usage rank#90
Requests / day188.7K
Tool calls / req0.48
Per request · 30 days
Avg input53.2K3.6× median
Avg output1.3K2.0× median
Cache hit75%about median
Reasoning share55%about median
Tool-call failures1.2%0.6× median

Consumption

16.7B33.3B50.0B3 Oct17 Oct1 Nov
Peak 48.3B 8 OctLow 6.7B 1 NovRank today #90

News & trends

11 mentions this week ▼ 21%this model · videos, posts, articles · 30 days
Trending · 14 days · Claude Opus 4 family
News · latest from Anthropic

Benchmarks

bar = share of the top score · Benchmark data: Artificial Analysis
GPQA Diamond#23 of 445
92.0%
τ²-Bench#27 of 319
94.4%
Terminal-Bench 2.1#13 of 182
84.6%
Coding index#15 of 196
74.3
Terminal-Bench Hard#5 of 311
58.3%
Long-context reasoning#82 of 393
77.7%
SciCode#36 of 160
54.4%
Humanity's Last Exam#15 of 465
48.7%
Speed · median across 6 providers, latest day
Output speed#149/33472 tok/sabout median
Fastest provider133 tok/sbest of 6 providers
First token#1380.88 s1.2× quicker than median

#18 cheapest of 19 with a similar score

Value
Intelligence Index
253035404550$2$5$10Claude Opus 4.8
Blended price per 1M tokens (log)
Family progress
Intelligence Index
2530354045Dec2026FebMarAprMayClaude Opus 4.8
  • Family releases41.8

Technical details

Reasoning
on · high
Tokenizer
Claude
Moderation
moderated
Reasoning effort
  • max
  • xhigh
  • high · default
  • medium
  • low
  • none
Request parameters 11
  • tools
  • tool_choice
  • response_format
  • structured_outputs
  • reasoning
  • reasoning_effort
  • include_reasoning
  • max_tokens
  • max_completion_tokens
  • stop
+1 more
  • verbosity

Interesting facts

  1. Less output room than similar models (128.0K against a 289.0K median).
  2. The most used of 5 Claude Opus 4 models by tokens last month.
  3. Weekdays average 32.9B tokens a day against 16.4B on weekends.
  4. Released 158 days ago, the newest of 4 Claude Opus 4 models.

When to use

Best for

  • Browser automation agents
  • Expert reasoning and science questions
  • Coding agents in terminal environments
  • Long document and image analysis

Think twice for

  • Multi-hour autonomous agent runs (weakest suite)
  • Data analysis
  • High-volume simple chat (cheaper options exist)
  • Self-hosting (closed weights)

Research papers

Same family

Claude Opus 4

Alternatives

other vendors · similar capability

Want Claude Opus 4.8 on OSMEXO?

Sign in to request it. Requests are reviewed; a request does not guarantee availability.

Request this model on OSMEXO