VerdictPal · editorial desk · 2026VerdictPal

Compare · Two models, head-to-head

Pick two models. See the diff.

Composite scores, benchmark ranks, capability grids, and pricing per million tokens. All sourced from the same data that powers the model atlas.

Cards Models

Claude Fable 5 vs Claude Opus 4.8

Model A

Model B

MODEL A

Claude Fable 5

Anthropic · Claude Fable

Composite 72Intel 7514 rows4 sources

Anthropic's June 2026 Mythos-class GA release. #1 on the Vals Index (75.15%) and Vals Multimodal Index (74.15%). 1M context, 128K output, safety classifiers route high-risk queries to Opus 4.8. $10 / $50 per 1M tokens.

MODEL B

Claude Opus 4.8

Anthropic · Claude Opus

Composite 73Intel 7026 rows12 sources

Anthropic's May 2026 flagship. Tops the Vals Index (70.17%) and the Artificial Analysis Intelligence Index (61.4). SWE-bench Verified 88.6%, Terminal-Bench 2.0 74.6%, 1M-token context, $15 / $75 per 1M tokens.

Claude Fable 53stats won

Tied2stats

Claude Opus 4.82stats won

Spec showdown

Capacity, price, freshness.

Claude Fable 5vsClaude Opus 4.8

Composite scores are within 5 points — 72 for Claude Fable 5 vs 73 for Claude Opus 4.8. The choice comes down to capability profile and price. Claude Fable 5 has the higher Vals/AA intelligence score (75 vs 70). Claude Fable 5 is more affordable at $10/1M input — about 2× cheaper than $15/1M.

Composite score
Claude Fable 572Claude Opus 4.873Winner on this row
Intelligence (Vals/AA)
Claude Fable 575Winner on this rowClaude Opus 4.870
Context window
Claude Fable 51MClaude Opus 4.81M
Max output
Claude Fable 5128KClaude Opus 4.8128K
Input price / 1M (lower wins)
Claude Fable 5$10Winner on this rowClaude Opus 4.8$15
Output speed (AA default API)
Claude Fable 570 t/sWinner on this rowClaude Opus 4.863 t/s
Time to first token (AA default API)
Claude Fable 568.33sClaude Opus 4.830.13sWinner on this row

Composite score

Intelligence (Vals/AA)

Context window

Max output

128K

Input price / 1M (lower wins)

$10

$15

Output speed (AA default API)

70 t/s

63 t/s

Time to first token (AA default API)

68.33s

30.13s

Capability grid

What each model can do.

Modalities

Claude Fable 5Claude Opus 4.8

Text●●

Image●—

Audio——

Video——

Code●●

Tool use●●

Reasoning●●

Access

Claude Fable 5Claude Opus 4.8

API●●

Consumer app●●

Open weights——

Managed cloud●●

On-device——

Family radar

Top score per benchmark family.

Reasoning

Coding

Agentic

100

Long context

Benchmark ranks

Every shared benchmark, side by side.

Reasoning

Artificial Analysis Intelligence Index#1/67 · #4/67

59.9%

55.7%

GPQA Diamond#8/69 · #51/69

92.6%

79.8%

Humanity's Last Exam#1/65 · #3/65

53.3%

45.7%

IFBench#39/57 · #41/57

63.47%

62.24%

Coding

DeepSWE#1/6 · #3/6

70%

59%

SciCode#1/65 · #14/65

60.2%

53.5%

Agentic

GDPval#1/6 · #2/6

1818

1638

Terminal-Bench#4/69 · #17/69

84.64%

74.6%

Vals Index#1/20 · #5/20

75.1%

70.4%

τ³-Bench Banking#15/64 · #30/64

62.4%

27.63%

Long context

AA-LCR#22/64 · #48/64

70%

58.3%

Performance

AA output speed#40/69 · #45/69

70 t/s

63 t/s

AA time to first token#7/69 · #13/69

68.33s

30.13s

Dossier facts

Dates, sources, fineprint.

Claude Fable 5Claude Opus 4.8

ProviderAnthropicAnthropic

FamilyClaude FableClaude Opus

Release date2026-06-092026-05-28

Knowledge cutoff2026-04-012026-02-01

Context window1M1M

Max output128K128K

Input price /1M$10$15

Output price /1M$50$75

Statusflagshipflagship

Composite score7273

Intelligence score7570

Intel sourcevalsvals

Benchmark rows1426

Source count412

Freshnessfreshfresh

Pricing checked2026-06-092026-06-01

Ecosystem

Which tools wrap each model.

Claude Fable 5

Claude Opus 4.8

Verdict

Which model to actually pick.

Open Claude Fable 5's dossier Open Claude Opus 4.8's dossier