Skip to main content
Radar

Five or fewer confirmed AI changes, with original sources, on mornings when something changed.A free source-linked morning brief.

Start free brief
BenchLM
Model A
Claude Opus 4.8

Anthropic

77.0/100

Supported · Public rank #7

90% interval 74.3–79.6

Claude Opus 4.8 vs Sakana Fugu

Updated August 13, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

Model B
Sakana Fugu

Sakana AI

—

Evidence status unavailable

90% interval unavailable

Decision reading

At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.

7 results are shared. Category rows based on different benchmark sets are marked directional and do not name a winner.

Which one for your work

Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.

No workload recommendation clears the current evidence threshold.

Use the matched evidence, workload costs, and sourced specifications below instead of treating a point score as a universal answer.

Show secondary and unsupported calls

  • Coding work

    Code generation, repair, and software-engineering tasks

    Not enough matched evidence

    The category averages use different weighted benchmark sets, so they are directional rather than like-for-like.

    Confidence: limited

  • Agentic work

    Tool use, computer use, and multi-step task completion

    Not enough matched evidence

    The category averages use different weighted benchmark sets, so they are directional rather than like-for-like.

    Confidence: limited

  • Long documents

    Prompts that approach the documented context limit

    No clear pick

    The documented context windows are equal.

    Confidence: documented

  • Chat turn cost

    1K fresh input + 500 output tokens

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: listed-rates

  • Cache-heavy agent loop cost

    200K cached + 20K fresh input + 10K output tokens

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: rate-fallback

  • Repository review cost

    50K fresh input + 3K output tokens

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: listed-rates

What is actually comparable

Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.

Shared results
7
Claude Opus 4.8 only
33
Sakana Fugu only
4
Like-for-like categories
0 / 8

4 categories use different evidence sets. Those rows remain visible for coverage context but do not name a winner.

Category results, on a stated basis

Each row states whether both averages use the same weighted benchmark set. Directional and not-comparable rows remain visible, but they never receive a winner in this template.

Agentic

Directional only
Claude Opus 4.8
80.3
Sakana Fugu
80.2
Weighted basis
3 vs 1 rows
Reading
Directional only

Coding

Directional only
Claude Opus 4.8
81.1
Sakana Fugu
59.7
Weighted basis
2 vs 2 rows
Reading
Directional only

Knowledge

Directional only
Claude Opus 4.8
62.7
Sakana Fugu
95.5
Weighted basis
2 vs 1 rows
Reading
Directional only

Multimodal

Directional only
Claude Opus 4.8
77.0
Sakana Fugu
85.1
Weighted basis
2 vs 1 rows
Reading
Directional only

Reasoning

Not comparable
Claude Opus 4.8
72.1
Sakana Fugu
86.6
Weighted basis
1 vs 1 rows
Reading
Not comparable

Math

Not comparable
Claude Opus 4.8
53.9
Sakana Fugu
Not measured
Weighted basis
3 vs 0 rows
Reading
Not comparable

Multilingual

Not comparable
Claude Opus 4.8
Not measured
Sakana Fugu
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Instruction following

Not comparable
Claude Opus 4.8
Not measured
Sakana Fugu
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Shape of the matched evidence

Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.

Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.

What each workload costs

Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.

Chat turn

1K fresh input + 500 output tokens

Claude Opus 4.8
$0.0175
Fits in one request
Sakana Fugu
API rate not published
Fits in one request

Sakana Fugu has no comparable published API token rate.

Repository review

50K fresh input + 3K output tokens

Claude Opus 4.8
$0.325
Fits in one request
Sakana Fugu
API rate not published
Fits in one request

Sakana Fugu has no comparable published API token rate.

Cache-heavy agent loop

200K cached + 20K fresh input + 10K output tokens

Claude Opus 4.8
$1.35
Fits in one request
Cached input priced at the published list-input rate
Sakana Fugu
API rate not published
Fits in one request
Cached-input rate unavailable

Claude Opus 4.8 has no published cached-input rate, so cached tokens use its listed input rate. Sakana Fugu has no comparable published API token rate.

Specification differences

Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.

Context window

Maximum documented context; output-token limits may be lower.

Claude Opus 4.8

1M

Anthropic model overview

Sakana Fugu

1M

Cached-input rate

A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.

Claude Opus 4.8

Not published

Sakana Fugu

No comparable hosted API rate

Reasoning profile

Claude Opus 4.8

Reasoning

Sakana Fugu

Reasoning

Weight access

Claude Opus 4.8

Proprietary

Sakana Fugu

Proprietary

License

Claude Opus 4.8

Proprietary

Sakana Fugu

Proprietary

Release date

Claude Opus 4.8

2026-05-28

Sakana Fugu

2026-06-22

If you already use one of these models
Deployment change
The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
Quality signal
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.
Workload cost
A complete comparable API-rate estimate is not available for both models.
Context tradeoff
Both models list 1M.

Run the same representative tasks against both endpoints before changing production traffic.

Benchmark evidence

The full public result ledger is available for audit without forcing a wide desktop table onto a phone.

Browse raw public benchmark evidence44 rows

Agentic

  • Terminal-Bench 3.0

    Claude Opus 4.821.1%
    Source
    Sakana Fugu—

    Not directly comparable

  • Terminal-Bench 2.0

    Claude Opus 4.874.6%
    Source
    Sakana Fugu80.2%
    Source

    Sakana Fugu leads this result

  • BrowseComp

    Claude Opus 4.884.3%
    Source
    Sakana Fugu—

    Not directly comparable

  • DeepSearchQA

    Claude Opus 4.893.1%
    Source
    Sakana Fugu—

    Not directly comparable

  • OSWorld-Verified

    Claude Opus 4.883.4%
    Source
    Sakana Fugu—

    Not directly comparable

  • Finance Agent v2

    Claude Opus 4.853.9%
    Source
    Sakana Fugu—

    Not directly comparable

  • MCP Atlas

    Claude Opus 4.882.2%
    Source
    Sakana Fugu—

    Not directly comparable

  • Toolathlon

    Claude Opus 4.859.9%
    Source
    Sakana Fugu—

    Not directly comparable

  • Gert Labs

    Claude Opus 4.872.97%
    Source
    Sakana Fugu—

    Not directly comparable

  • ResearchClawBench

    Claude Opus 4.821.1%
    Source
    Sakana Fugu—

    Not directly comparable

  • OSWorld 2.0

    Claude Opus 4.820.6%
    Source
    Sakana Fugu—

    Not directly comparable

Coding

  • SWE-bench Verified

    Claude Opus 4.888.6%
    Source
    Sakana Fugu—

    Not directly comparable

  • SWE-bench Pro

    Claude Opus 4.869.2%
    Source
    Sakana Fugu59%
    Source

    Claude Opus 4.8 leads this result

  • SWE Multilingual

    Claude Opus 4.884.4%
    Source
    Sakana Fugu—

    Not directly comparable

  • SWE Multimodal

    Claude Opus 4.838.4%
    Source
    Sakana Fugu—

    Not directly comparable

  • Terminal-Bench 2.0

    Claude Opus 4.874.6%
    Source
    Sakana Fugu80.2%
    Source

    Sakana Fugu leads this result

  • cursorBench31

    Claude Opus 4.858.4%
    Source
    Sakana Fugu—

    Not directly comparable

  • cursorBench32

    Claude Opus 4.862.3%
    Source
    Sakana Fugu—

    Not directly comparable

  • FrontierCode 1.1 Main

    Claude Opus 4.846.5%
    Source
    Sakana Fugu—

    Not directly comparable

  • APEX-SWE

    Claude Opus 4.843.9%
    Source
    Sakana Fugu—

    Not directly comparable

  • EEBench

    Claude Opus 4.851.4%
    Source
    Sakana Fugu—

    Not directly comparable

  • 3DCodeBench

    Claude Opus 4.847.0%
    Source
    Sakana Fugu—

    Not directly comparable

  • CADGenBench Generation

    Claude Opus 4.827.4%
    Source
    Sakana Fugu—

    Not directly comparable

  • SpaceXAI MTS Eval

    Claude Opus 4.855.8%
    Source
    Sakana Fugu—

    Not directly comparable

  • InferenceEval

    Claude Opus 4.840.7%
    Source
    Sakana Fugu—

    Not directly comparable

  • KernelBench Internal

    Claude Opus 4.867.6%
    Source
    Sakana Fugu—

    Not directly comparable

  • LiveCodeBench v6

    Claude Opus 4.8—
    Sakana Fugu92.9%
    Source

    Not directly comparable

  • LiveCodeBench Pro

    Claude Opus 4.8—
    Sakana Fugu87.8%
    Source

    Not directly comparable

  • SciCode

    Claude Opus 4.8—
    Sakana Fugu60.1%
    Source

    Not directly comparable

Reasoning

  • ARC-AGI-2

    Claude Opus 4.872.1%
    Source
    Sakana Fugu—

    Not directly comparable

  • ARC-AGI-3

    Claude Opus 4.81.5%
    Source
    Sakana Fugu—

    Not directly comparable

  • MRCRv2

    Claude Opus 4.8—
    Sakana Fugu86.6%
    Source

    Not directly comparable

Knowledge

  • GPQA

    Claude Opus 4.893.6%
    Source
    Sakana Fugu95.5%
    Source

    Sakana Fugu leads this result

  • GPQA-D

    Claude Opus 4.893.6%
    Source
    Sakana Fugu95.5%
    Source

    Sakana Fugu leads this result

  • HLE

    Claude Opus 4.857.9%
    Source
    Sakana Fugu—

    Not directly comparable

  • HLE w/o tools

    Claude Opus 4.849.8%
    Source
    Sakana Fugu47.2%
    Source

    Claude Opus 4.8 leads this result

Math

  • USAMO 2026

    Claude Opus 4.896.7%
    Source
    Sakana Fugu—

    Not directly comparable

  • FrontierMath v2 (Tiers 1-3)

    Claude Opus 4.847.241%
    Source
    Sakana Fugu—

    Not directly comparable

  • FrontierMath v2 (Tier 4)

    Claude Opus 4.831.250%
    Source
    Sakana Fugu—

    Not directly comparable

Multilingual

  • INCLUDE

    Claude Opus 4.887.6%
    Source
    Sakana Fugu—

    Not directly comparable

Multimodal

  • OfficeQA Pro

    Claude Opus 4.866.2%
    Source
    Sakana Fugu—

    Not directly comparable

  • ScreenSpot Pro

    Claude Opus 4.887.9%
    Source
    Sakana Fugu—

    Not directly comparable

  • CharXiv

    Claude Opus 4.889.9%
    Source
    Sakana Fugu85.1%
    Source

    Claude Opus 4.8 leads this result

  • CharXiv w/o tools

    Claude Opus 4.880.5%
    Source
    Sakana Fugu—

    Not directly comparable

Frequently asked questions

Which is better, Claude Opus 4.8 or Sakana Fugu?

At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner. The page therefore keeps the decision tied to the specific documented workload.

Which is better for coding, Claude Opus 4.8 or Sakana Fugu?

The current coding averages use different weighted benchmark sets, so BenchLM does not name a winner from them. Read the shared benchmark rows directly and test the models on the same task set.

Which is better for agentic tasks, Claude Opus 4.8 or Sakana Fugu?

The current agentic tasks averages use different weighted benchmark sets, so BenchLM does not name a winner from them. Read the shared benchmark rows directly and test the models on the same task set.

Which costs less, Claude Opus 4.8 or Sakana Fugu?

Both models do not have comparable published API token rates, so this page does not name a universal price winner.

Which has the larger context window, Claude Opus 4.8 or Sakana Fugu?

Both models list the same context window, 1M.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated August 13, 2026

Watch Claude Opus 4.8 vs Sakana Fugu

One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.

Read a sample issue

Join 2,000+ readers.