ZMIME
Comparison · 3 models · Updated Nov 18, 2026

Gemma 4 E4B IT vs MiMo-V2.5-Pro-UltraSpeed vs Step 3.5 Flash 2603

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

Price in rating: Excluded · Include price

  1. Google

    Gemma 4 E4B IT

    Released Apr 2, 2026

    55/100
    • ECI—
    • Price$0.085 / $0.35
    • Context131K
  2. Xiaomi

    MiMo-V2.5-Pro-UltraSpeed

    Released Jun 8, 2026Deprecated

    47/100
    • ECI—
    • Price$1.30 / $2.61
    • Context1.05M
  3. StepFun

    Step 3.5 Flash 2603

    Released Apr 2, 2026

    44/100
    • ECI—
    • Price$0.10 / $0.30
    • Context256K
01 — Verdict

Capability evidence is incomplete

There is not enough shared capability evidence to name an overall winner. The scores (Gemma 4 E4B IT 55/100, MiMo-V2.5-Pro-UltraSpeed 47/100, Step 3.5 Flash 2603 44/100) compare the available specifications only. 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
  • Lowest priceStep 3.5 Flash 2603Step 3.5 Flash 2603 $0.15 · Gemma 4 E4B IT $0.151 · MiMo-V2.5-Pro-UltraSpeed $1.63 per 1M tokens (3:1 blend)
  • Longest contextMiMo-V2.5-Pro-UltraSpeedMiMo-V2.5-Pro-UltraSpeed 1,048,576 · Step 3.5 Flash 2603 256,000 · Gemma 4 E4B IT 131,072 tokens
  • Widest inputsGemma 4 E4B ITGemma 4 E4B IT: Text, Images, Audio · MiMo-V2.5-Pro-UltraSpeed: Text · Step 3.5 Flash 2603: Text
  • Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
How the score is built
MeasureWeightGemma 4 E4B ITMiMo-V2.5-Pro-UltraSpeedStep 3.5 Flash 2603
Inputs & features60%703545
Context window40%326543
Specs only100%55/10047/10044/100

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.

Gemma 4 E4B IT vs MiMo-V2.5-Pro-UltraSpeed vs Step 3.5 Flash 2603 specifications side by side
SpecificationGemma 4 E4B ITGoogleMiMo-V2.5-Pro-UltraSpeedXiaomiStep 3.5 Flash 2603StepFun
Capability
Capabilities Index (ECI)———
ECI rank———
Price per million tokens
Input$0.085 (best)$1.30$0.10
Output$0.35$2.61$0.30 (best)
Cached input—$0.011 (best)$0.02
Blended (3:1)$0.151$1.63$0.15 (best)
Long-context rateSame rateSame rateSame rate
Price sourceMedian of 2 providersOfficial Xiaomi APIOfficial StepFun (Global) API
Limits
Context window131,072 tokens1,048,576 tokens (best)256,000 tokens
Max output8,192 tokens131,072 tokens256,000 tokens (best)
Inputs and features
TextYesYesYes
ImagesYesNoNo
PDFsNoNoNo
AudioYesNoNo
VideoNoNoNo
ReasoningYesYesYeslow · high
Tool callingYesYesYes
Structured outputNoNoYes
Availability
WeightsOpenOpenOpen
API model ID—mimo-v2.5-pro-ultraspeedstep-3.5-flash-2603
API providers3 (best)13 (best)
ReleasedApr 2, 2026Jun 8, 2026Apr 2, 2026
Knowledge cutoff—Dec 2024Jan 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.

  • Gemma 4 E4B IT$1.55
  • MiMo-V2.5-Pro-UltraSpeed$18.27
  • Step 3.5 Flash 2603$1.60
04 — Questions

Which should you choose?

Which is better: Gemma 4 E4B IT, MiMo-V2.5-Pro-UltraSpeed or Step 3.5 Flash 2603?

There is not enough shared capability evidence to name an overall winner. The scores (Gemma 4 E4B IT 55/100, MiMo-V2.5-Pro-UltraSpeed 47/100, Step 3.5 Flash 2603 44/100) compare the available specifications only. 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, Gemma 4 E4B IT, MiMo-V2.5-Pro-UltraSpeed or Step 3.5 Flash 2603?

Step 3.5 Flash 2603 is cheaper at $0.10 input / $0.30 output per million tokens (official StepFun (Global) API price). Gemma 4 E4B IT costs $0.085 input / $0.35 output per million tokens (median across 2 API providers); MiMo-V2.5-Pro-UltraSpeed costs $1.30 input / $2.61 output per million tokens (official Xiaomi API price). At a typical mix of three input tokens to one output token, that is $0.15 per million tokens for Step 3.5 Flash 2603 versus $0.151 for Gemma 4 E4B IT (1× as much) and $1.63 for MiMo-V2.5-Pro-UltraSpeed (11× as much).

Which scores higher on benchmarks?

There is no independent benchmark that covers all three models yet. Gemma 4 E4B IT has not been scored yet, MiMo-V2.5-Pro-UltraSpeed has not been scored yet and Step 3.5 Flash 2603 has not been scored yet.

Which is better for coding?

There are no published SWE-bench Verified results for Gemma 4 E4B IT, MiMo-V2.5-Pro-UltraSpeed and Step 3.5 Flash 2603 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?

MiMo-V2.5-Pro-UltraSpeed has the largest context window at 1,048,576 tokens, against 256,000 for Step 3.5 Flash 2603 and 131,072 for Gemma 4 E4B IT. Maximum output per response: Gemma 4 E4B IT up to 8,192, MiMo-V2.5-Pro-UltraSpeed up to 131,072, Step 3.5 Flash 2603 up to 256,000 tokens.

Which can read images, PDFs, audio or video?

Gemma 4 E4B IT accepts text, images and audio; MiMo-V2.5-Pro-UltraSpeed accepts text; Step 3.5 Flash 2603 accepts text. Gemma 4 E4B IT handles the widest range of inputs.

Are any of these open source?

Yes, all three publish their weights, so you can self-host them.

Which is newer?

MiMo-V2.5-Pro-UltraSpeed is the newest, released Jun 8, 2026. Gemma 4 E4B IT came out Apr 2, 2026; Step 3.5 Flash 2603 came out Apr 2, 2026. Knowledge cutoff: MiMo-V2.5-Pro-UltraSpeed Dec 2024, Step 3.5 Flash 2603 Jan 2025.

How do you decide the winner?

This rating uses inputs & features (60%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (40%, 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.