ZMIME
Comparison · 3 models · Updated Nov 19, 2026

Fugu Ultra v2 vs Claude Mythos 5.2 vs Qwen 3.8 Max Prime

No independent capability benchmark covers every model here yet. The score uses price.

Price in rating: Included · Exclude price

  1. Sakana AI

    Fugu Ultra v2

    Released Sep 11, 2026

    0/100
    • ECI—
    • Price$5.00 / $30.00
    • Context1M
  2. Anthropic

    Claude Mythos 5.2

    Released Nov 9, 2026New

    0/100
    • ECI—
    • Price$8.00 / $40.00
    • Context—
  3. Alibaba (Qwen)

    Qwen 3.8 Max Prime

    Released Sep 23, 2026New

    12/100
    • ECI—
    • Price$4.00 / $12.00
    • Context1M
01 — Verdict

Capability evidence is incomplete

There is not enough shared capability evidence to name an overall winner. The scores (Qwen 3.8 Max Prime 12/100, Fugu Ultra v2 0/100, Claude Mythos 5.2 0/100) compare the available specifications and price. The score weighs price 100%. 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 priceQwen 3.8 Max PrimeQwen 3.8 Max Prime $6.00 · Fugu Ultra v2 $11.25 · Claude Mythos 5.2 $16.00 per 1M tokens (3:1 blend)
  • Longest contextFugu Ultra v2 and Qwen 3.8 Max PrimeFugu Ultra v2 1,000,000 · Qwen 3.8 Max Prime 1,000,000 tokens
  • Widest inputsFugu Ultra v2 and Qwen 3.8 Max PrimeFugu Ultra v2: Text, Images, PDFs · Claude Mythos 5.2: Text, Images · Qwen 3.8 Max Prime: Text, Images, Video
  • Self-hostingNo confirmed open weightsAll three are available only through APIs
How the score is built
MeasureWeightFugu Ultra v2Claude Mythos 5.2Qwen 3.8 Max Prime
Price100%0012
Specs & price only100%0/1000/10012/100

Left out because at least one model lacks the data: capability, inputs & features and context window. 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 v2 vs Claude Mythos 5.2 vs Qwen 3.8 Max Prime specifications side by side
SpecificationFugu Ultra v2Sakana AIClaude Mythos 5.2AnthropicQwen 3.8 Max PrimeAlibaba (Qwen)
Task—Restricted-access reasoning—
Published capabilitiesQuality-first multi-agent model for hard research, analysis, and competitionsShares its underlying model with Claude Fable 5.2. Access and safeguards differ.High-throughput edition of Qwen3.8 Max for coding, professional work, multimodal understanding, and long-running agent workflows
Access—Vetted organizations through Anthropic verification programs—
Sources—Publisher documentation—
Capability
Capabilities Index (ECI)———
ECI rank———
Price per million tokens
Input$5.00$8.00$4.00 (best)
Output$30.00$40.00$12.00 (best)
Cached input———
Blended (3:1)$11.25$16.00$6.00 (best)
Long-context rateSame rateSame rateSame rate
Price sourceMedian of 6 providersOfficial Anthropic APIMedian of 6 providers
Limits
Context window1,000,000 tokens—1,000,000 tokens
Max output128,000 tokens—131,072 tokens (best)
Inputs and features
TextYesYesYes
ImagesYesYesYes
PDFsYesNoNo
AudioNoNoNo
VideoNoNoYes
ReasoningYesUnconfirmedYes
Tool callingYes—Yes
Structured outputNo—No
Availability
WeightsProprietaryProprietaryProprietary
API model ID———
API providers6—6
ReleasedSep 11, 2026Nov 9, 2026Sep 23, 2026
Knowledge cutoffAug 28, 2026——
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 v2$110.00
  • Claude Mythos 5.2$160.00
  • Qwen 3.8 Max Prime$64.00
04 — Questions

Which should you choose?

Which is better: Fugu Ultra v2, Claude Mythos 5.2 or Qwen 3.8 Max Prime?

There is not enough shared capability evidence to name an overall winner. The scores (Qwen 3.8 Max Prime 12/100, Fugu Ultra v2 0/100, Claude Mythos 5.2 0/100) compare the available specifications and price. The score weighs price 100%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.

Which is cheaper, Fugu Ultra v2, Claude Mythos 5.2 or Qwen 3.8 Max Prime?

Qwen 3.8 Max Prime is cheaper at $4.00 input / $12.00 output per million tokens (median across 6 API providers). Fugu Ultra v2 costs $5.00 input / $30.00 output per million tokens (median across 6 API providers); Claude Mythos 5.2 costs $8.00 input / $40.00 output per million tokens (official Anthropic API price). At a typical mix of three input tokens to one output token, that is $6.00 per million tokens for Qwen 3.8 Max Prime versus $11.25 for Fugu Ultra v2 (1.9× as much) and $16.00 for Claude Mythos 5.2 (2.7× as much).

Which scores higher on benchmarks?

There is no independent benchmark that covers all three models yet. Fugu Ultra v2 has not been scored yet, Claude Mythos 5.2 has not been scored yet and Qwen 3.8 Max Prime has not been scored yet.

Which is better for coding?

There are no published SWE-bench Verified results for Fugu Ultra v2, Claude Mythos 5.2 and Qwen 3.8 Max Prime yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. Note that Claude Mythos 5.2 has no confirmed tool-calling support in this comparison.

Which has the bigger context window?

Published context limits: Fugu Ultra v2 1,000,000 tokens, Claude Mythos 5.2 unconfirmed, Qwen 3.8 Max Prime 1,000,000 tokens. Maximum output per response: Fugu Ultra v2 up to 128,000, Qwen 3.8 Max Prime up to 131,072 tokens.

Which can read images, PDFs, audio or video?

Fugu Ultra v2 accepts text, images and PDFs; Claude Mythos 5.2 accepts text and images; Qwen 3.8 Max Prime accepts text, images and video. Fugu Ultra v2 handles the widest range of inputs.

Are any of these open source?

No. Fugu Ultra v2, Claude Mythos 5.2 and Qwen 3.8 Max Prime are proprietary and only available through APIs and apps.

Which is newer?

Claude Mythos 5.2 is the newest, released Nov 9, 2026. Qwen 3.8 Max Prime came out Sep 23, 2026; Fugu Ultra v2 came out Sep 11, 2026. Knowledge cutoff: Fugu Ultra v2 Aug 28, 2026.

How do you decide the winner?

This rating uses price (100%, blended price per million tokens (3 input : 1 output), 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 19, 2026.