ZMIME
Comparison · 3 models · Updated Nov 18, 2026

Fugu Ultra vs Kimi K2.7 Code Highspeed vs Muse Wonder Full

No independent capability benchmark covers every model here yet. The score uses price, inputs & features and context window.

Price in rating: Included · Exclude price

  1. Sakana AI

    Fugu Ultra

    Released Jun 15, 2026

    34/100
    • ECI—
    • Price$5.00 / $30.00
    • Context1M
  2. Moonshot AI

    Kimi K2.7 Code Highspeed

    Released Jun 12, 2026

    41/100
    • ECI—
    • Price$1.90 / $8.00
    • Context262K
  3. Meta

    Muse Wonder Full

    Released Oct 12, 2026New

    39/100
    • ECI—
    • Price$4.80 / $24.00
    • Context1.05M
01 — Verdict

Capability evidence is incomplete

There is not enough shared capability evidence to name an overall winner. The scores (Kimi K2.7 Code Highspeed 41/100, Muse Wonder Full 39/100, Fugu Ultra 34/100) compare the available specifications and price. The score weighs price 50%, inputs & features 30%, context window 20%. 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
  • Lowest priceKimi K2.7 Code HighspeedKimi K2.7 Code Highspeed $3.42 · Muse Wonder Full $9.60 · Fugu Ultra $11.25 per 1M tokens (3:1 blend)
  • Longest contextMuse Wonder FullMuse Wonder Full 1,048,576 · Fugu Ultra 1,000,000 · Kimi K2.7 Code Highspeed 262,144 tokens
  • Widest inputsMuse Wonder FullFugu Ultra: Text, Images · Kimi K2.7 Code Highspeed: Text, Images, Video · Muse Wonder Full: Text, Images, Audio, Video
  • Self-hostingKimi K2.7 Code HighspeedPublishes downloadable weights
How the score is built
MeasureWeightFugu UltraKimi K2.7 Code HighspeedMuse Wonder Full
Price50%0233
Inputs & features30%707080
Context window20%644365
Specs & price only100%34/10041/10039/100

Left out because at least one model lacks the data: capability. 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.

Fugu Ultra vs Kimi K2.7 Code Highspeed vs Muse Wonder Full specifications side by side
SpecificationFugu UltraSakana AIKimi K2.7 Code HighspeedMoonshot AIMuse Wonder FullMeta
Capability
Capabilities Index (ECI)———
ECI rank———
Price per million tokens
Input$5.00$1.90 (best)$4.80
Output$30.00$8.00 (best)$24.00
Cached input$0.50——
Blended (3:1)$11.25$3.42 (best)$9.60
Long-context rateOver 272K: $10.00 / $45.00Same rateSame rate
Price sourceOfficial Sakana AI APIMedian of 12 providersMedian of 1 providers
Limits
Context window1,000,000 tokens262,144 tokens1,048,576 tokens (best)
Max output—262,144 tokens—
Inputs and features
TextYesYesYes
ImagesYesYesYes
PDFsNoNoNo
AudioNoNoYes
VideoNoYesYes
ReasoningYeshigh · xhighYesYes
Tool callingYesYesYes
Structured outputYesNoNo
Availability
WeightsProprietaryOpenProprietary
API model IDfugu-ultra——
API providers13 (best)121
ReleasedJun 15, 2026Jun 12, 2026Oct 12, 2026
Knowledge cutoff—Jan 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.

  • Fugu Ultra$110.00
  • Kimi K2.7 Code Highspeed$35.00
  • Muse Wonder Full$96.00
04 — Questions

Which should you choose?

Which is better: Fugu Ultra, Kimi K2.7 Code Highspeed or Muse Wonder Full?

There is not enough shared capability evidence to name an overall winner. The scores (Kimi K2.7 Code Highspeed 41/100, Muse Wonder Full 39/100, Fugu Ultra 34/100) compare the available specifications and price. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.

Which is cheaper, Fugu Ultra, Kimi K2.7 Code Highspeed or Muse Wonder Full?

Kimi K2.7 Code Highspeed is cheaper at $1.90 input / $8.00 output per million tokens (median across 12 API providers). Muse Wonder Full costs $4.80 input / $24.00 output per million tokens (median across 1 API provider); Fugu Ultra costs $5.00 input / $30.00 output per million tokens (official Sakana AI API price). At a typical mix of three input tokens to one output token, that is $3.42 per million tokens for Kimi K2.7 Code Highspeed versus $9.60 for Muse Wonder Full (2.8× as much) and $11.25 for Fugu Ultra (3.3× as much).

Which scores higher on benchmarks?

There is no independent benchmark that covers all three models yet. Fugu Ultra has not been scored yet, Kimi K2.7 Code Highspeed has not been scored yet and Muse Wonder Full has not been scored yet.

Which is better for coding?

There are no published SWE-bench Verified results for Fugu Ultra, Kimi K2.7 Code Highspeed and Muse Wonder Full yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. All three support tool calling for agent workflows.

Which has the bigger context window?

Muse Wonder Full has the largest context window at 1,048,576 tokens, against 1,000,000 for Fugu Ultra and 262,144 for Kimi K2.7 Code Highspeed. Maximum output per response: Kimi K2.7 Code Highspeed up to 262,144 tokens.

Which can read images, PDFs, audio or video?

Fugu Ultra accepts text and images; Kimi K2.7 Code Highspeed accepts text, images and video; Muse Wonder Full accepts text, images, audio and video. Muse Wonder Full handles the widest range of inputs.

Are any of these open source?

Kimi K2.7 Code Highspeed publishes its weights and can be self-hosted; Fugu Ultra and Muse Wonder Full is proprietary.

Which is newer?

Muse Wonder Full is the newest, released Oct 12, 2026. Fugu Ultra came out Jun 15, 2026; Kimi K2.7 Code Highspeed came out Jun 12, 2026. Knowledge cutoff: Kimi K2.7 Code Highspeed Jan 2025.

How do you decide the winner?

This rating uses price (50%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (30%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (20%, 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 Nov 18, 2026.