ZMIME
Comparison · 2 models · Updated Oct 4, 2026

DiffusionGemma 26B A4B IT vs Fugu

Fugu comes out ahead, 66 to 51 on our weighted score.

  1. Google

    DiffusionGemma 26B A4B IT

    Released Jun 9, 2026

    51/100
    • ECI—
    • Price—
    • Context262K
    Full specs Remove DiffusionGemma 26B A4B IT
  2. Our pick

    Sakana AI

    Fugu

    Released Jun 15, 2026

    66/100
    • ECI—
    • Price—
    • Context1M
    Full specs Remove Fugu

  3. Add a model

    Make it a three-way comparison.

01 — Verdict

Fugu is our pick

Fugu is the better all-round choice, scoring 66/100 against DiffusionGemma 26B A4B IT (51). It leads on inputs & features and context window. The score weighs inputs & features 60%, context window 40%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.

  • CapabilityNot enough dataNo independent benchmark covers every model here yet
  • Longest contextFuguFugu 1,000,000 · DiffusionGemma 26B A4B IT 262,144 tokens
  • Widest inputsSame inputsDiffusionGemma 26B A4B IT: Text, Images · Fugu: Text, Images
  • Self-hostingDiffusionGemma 26B A4B ITPublishes downloadable weights (Apache 2.0)
How the score is built
MeasureWeightDiffusionGemma 26B A4B ITFugu
Inputs & features60%6070
Context window40%3760
Overall100%51/10066/100

Left out because at least one model lacks the data: capability and price. The remaining weights were rescaled.

02 — Side by side

Every spec in one table

Highlighted cells lead their row. Dashes mean the data is not published.

DiffusionGemma 26B A4B IT vs Fugu specifications side by side
SpecificationDiffusionGemma 26B A4B ITGoogleFuguSakana AI
Capability
Capabilities Index (ECI)——
ECI rank——
Price per million tokens
Input——
Output——
Cached input——
Blended (3:1)——
Long-context rate——
Price source——
Limits
Context window262,144 tokens1,000,000 tokens (best)
Max output32,768 tokens—
Inputs and features
TextYesYes
ImagesYesYes
PDFsNoNo
AudioNoNo
VideoNoNo
ReasoningYesYeshigh · xhigh
Tool callingYesYes
Structured outputNoYes
Availability
WeightsOpenApache 2.0Proprietary
API model ID—fugu
API providers11
ReleasedJun 9, 2026Jun 15, 2026
Knowledge cutoffJan 2025—
03 — Cost

What would a month cost?

Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.

  • DiffusionGemma 26B A4B IT—
  • Fugu—
04 — Questions

Which should you choose?

Which is better: DiffusionGemma 26B A4B IT or Fugu?

Fugu is the better all-round choice, scoring 66/100 against DiffusionGemma 26B A4B IT (51). It leads on inputs & features and context window. The score weighs inputs & features 60%, context window 40%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.

Which is cheaper, DiffusionGemma 26B A4B IT or Fugu?

None of these models has a published per-token price.

Which scores higher on benchmarks?

There is no independent benchmark that covers both models yet. DiffusionGemma 26B A4B IT has not been scored yet and Fugu has not been scored yet.

Which is better for coding?

There are no published SWE-bench Verified results for DiffusionGemma 26B A4B IT and Fugu yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. Both support tool calling for agent workflows.

Which has the bigger context window?

Fugu has the largest context window at 1,000,000 tokens, against 262,144 for DiffusionGemma 26B A4B IT. Maximum output per response: DiffusionGemma 26B A4B IT up to 32,768 tokens.

Which can read images, PDFs, audio or video?

DiffusionGemma 26B A4B IT accepts text and images; Fugu accepts text and images. They handle the same number of input types.

Are any of these open source?

DiffusionGemma 26B A4B IT publishes its weights (Apache 2.0) and can be self-hosted; Fugu is proprietary.

Which is newer?

Fugu is the newest, released Jun 15, 2026. DiffusionGemma 26B A4B IT came out Jun 9, 2026. Knowledge cutoff: DiffusionGemma 26B A4B IT Jan 2025.

How do you decide the winner?

Each model gets a 0–100 score on capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Oct 4, 2026.

05 — Keep comparing

Related comparisons