Advertise here
#179

Claude Opus 4.8

Anthropic Release: 2026-05-28 Tested on: 2026-10-01 09:25 anthropic/claude-opus-4.8::none
~5T total (~500B active)MoEClosedEstimated
(medium) (low) (none)

Summary

Claude Opus 4.8 scores 6.9 on AI BENCHY and ranks #179. It has 10.0 reliability, a 60.9% pass rate, $2.334 total cost, and 7.54s average response time.

What makes Claude Opus 4.8 unique: It stands out most in Combined, where it ranks #3, while Data parsing and extraction is its weakest area at #17.

Model facts

Researched on 2026-08-12

Estimated
Parameters
~5T total (~500B active)
Architecture
MoE
Availability
Closed
License
-

Best estimate from public evidence; the vendor did not disclose every value. External industry estimate; vendor has not disclosed the architecture.

Parameter source Availability source

Score

6.9

Consistency

9.3

Reliability

10.0

Total Cost (Current Price)

$2.334

Total Output Tokens

24,343

Total Input Tokens

345,059

Input Price

$5.000 / 1M

Output Price

$25.000 / 1M

Cache Read Price

$0.500 / 1M

Cache Write Price

$6.250 / 1M

Cache prices apply to input tokens. Reads reuse cached prompts; writes store them and can cost extra. Output tokens use the output price.

Tests Correct

Wrong Tests: 10

Attempt pass rate: 60.9%

Flaky tests

2

Flaky tests had mixed outcomes across runs (at least one pass and one fail).

Response Time (avg)

7.54s

Response Time (max): 65.07s

Response Time (total): 173.31s

Wrong answer: 5 Extra formatting: 3 Did not follow instructions: 1 No answer: 1

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#179 Claude Opus 4.8

none
Cost
$0.053
Time
22.0s
Tokens
2,253 tok

Run history

Tested on Score Reliability Tests Correct Total Cost Compare
2026-10-01 09:25 Re-test 6.9 10.0 $2.334 Current run
2026-07-16 22:29 New test added 7.3 10.0 $1.166 Compare
2026-06-30 22:01 Re-test 7.2 10.0 $0.539 Compare
2026-06-04 13:44 New test added 7.0 10.0 $0.539 Compare
2026-05-28 17:34 Initial run 7.3 10.0 $0.519 Compare

This run used a different benchmark suite. Keep suite changes in mind when reading historical movement.

Run comparison

RunBenchmark coverageScoreConsistencyReliabilityTests CorrectFlaky testsTotal Output TokensTotal Input TokensTotal CostResponse Time (avg)
2026-10-01 09:25 · Current run69/69 attempts6.99.310.013/23224,343345,059$2.3347.54s
2026-07-16 22:29 · New test added66/66 attempts7.39.210.013/22216,797149,206$1.1664.91s
Difference—-0.4+0.10.000+7546+195853+$1.168+2625ms

Benchmark coverage differs: 69/69 attempts (Target: 3 repeats per test) versus 66/66 attempts (Target: 3 repeats per test). Totals and repeat-sensitive metrics are not directly comparable.

These two runs used different benchmark suites, so the deltas reflect both model changes and suite changes.

Price History

Historical pricing data for this model from OpenRouter.

Date Input Price Output Price Cache Read Price Cache Write Price
2026-08-14 12:51 $5.000 / 1M $25.000 / 1M N/A N/A
2026-10-01 18:33 $5.000 / 1M $25.000 / 1M $0.500 / 1M $6.250 / 1M

Charts

Choose the first model, then click a second model to open a side-by-side page.

Total Output Tokens

Score vs Total Output Tokens

Quick Compare

Claude Opus 4.8nonevsClaude Sonnet 5.5lowClaude Opus 4.8nonevsKimi K2.6mediumClaude Opus 4.8nonevsGPT-5.6 LunalowClaude Opus 4.8nonevsQwen3.5-FlashmediumClaude Opus 4.8nonevsGemma 4 31BmediumFree AvailableClaude Opus 4.8nonevsInklinglowFree AvailableClaude Opus 4.8nonevsEmber-1highClaude Opus 4.8nonevsMercury 2.5 PreviewmediumClaude Opus 4.8nonevsGemini 3.1 Flash LitemediumClaude Opus 4.8nonevsGPT-6 Solnone

Category Breakdown

Category Score Consistency Tests Correct
Agentic 5.0 10.0
Anti-AI Tricks 6.5 10.0
Coding 5.5 10.0
Combined 9.8 10.0
Data parsing and extraction 7.3 5.8
Domain specific 5.3 7.2
General Intelligence 10.0 10.0
Instructions following 9.9 10.0
Puzzle Solving 7.7 10.0
Tool Calling 10.0 10.0
Trivia 3.0 10.0

Compared models