ZMIME
Comparison · 3 models · Updated Nov 18, 2026

Gemini 3.1 Flash Lite vs Macaron V1 Tall vs MiMo-V2.5-Pro

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

    Gemini 3.1 Flash Lite

    Released May 7, 2026

    86/100
    • ECI—
    • Price$0.25 / $1.50
    • Context1.05M
  2. Mindlab

    Macaron V1 Tall

    Released Jul 21, 2026

    53/100
    • ECI—
    • Price$0.247 / $1.43
    • Context262K
  3. Xiaomi

    MiMo-V2.5-Pro

    Released Apr 22, 2026Deprecated

    47/100
    • ECI—
    • Price$0.435 / $0.87
    • Context1.05M
01 — Verdict

Capability evidence is incomplete

There is not enough shared capability evidence to name an overall winner. The scores (Gemini 3.1 Flash Lite 86/100, Macaron V1 Tall 53/100, MiMo-V2.5-Pro 47/100) 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 priceMacaron V1 TallMacaron V1 Tall $0.543 · MiMo-V2.5-Pro $0.544 · Gemini 3.1 Flash Lite $0.563 per 1M tokens (3:1 blend)
  • Longest contextGemini 3.1 Flash Lite and MiMo-V2.5-ProGemini 3.1 Flash Lite 1,048,576 · MiMo-V2.5-Pro 1,048,576 · Macaron V1 Tall 262,144 tokens
  • Widest inputsGemini 3.1 Flash LiteGemini 3.1 Flash Lite: Text, Images, PDFs, Audio, Video · Macaron V1 Tall: Text, Images · MiMo-V2.5-Pro: Text
  • Self-hostingMacaron V1 Tall and MiMo-V2.5-ProPublishes downloadable weights (MIT)
How the score is built
MeasureWeightGemini 3.1 Flash LiteMacaron V1 TallMiMo-V2.5-Pro
Inputs & features60%1006035
Context window40%654365
Specs only100%86/10053/10047/100

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.

Gemini 3.1 Flash Lite vs Macaron V1 Tall vs MiMo-V2.5-Pro specifications side by side
SpecificationGemini 3.1 Flash LiteGoogleMacaron V1 TallMindlabMiMo-V2.5-ProXiaomi
Capability
Capabilities Index (ECI)———
ECI rank———
Price per million tokens
Input$0.25$0.247 (best)$0.435
Output$1.50$1.43$0.87 (best)
Cached input$0.025—$0.0036 (best)
Blended (3:1)$0.563$0.543 (best)$0.544
Long-context rateSame rateSame rateSame rate
Price sourceOfficial Google APIMedian of 1 providersOfficial Xiaomi API
Limits
Context window1,048,576 tokens (best)262,144 tokens1,048,576 tokens (best)
Max output65,536 tokens32,768 tokens131,072 tokens (best)
Inputs and features
TextYesYesYes
ImagesYesYesNo
PDFsYesNoNo
AudioYesNoNo
VideoYesNoNo
ReasoningYesminimal · low · medium · highYesYes
Tool callingYesYesYes
Structured outputYesNoNo
Availability
WeightsProprietaryOpenMITOpen
API model IDgemini-3.1-flash-lite—mimo-v2.5-pro
API providers26 (best)115
ReleasedMay 7, 2026Jul 21, 2026Apr 22, 2026
Knowledge cutoffJan 2025—Dec 2024
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.

  • Gemini 3.1 Flash Lite$5.50
  • Macaron V1 Tall$5.33
  • MiMo-V2.5-Pro$6.09
04 — Questions

Which should you choose?

Which is better: Gemini 3.1 Flash Lite, Macaron V1 Tall or MiMo-V2.5-Pro?

There is not enough shared capability evidence to name an overall winner. The scores (Gemini 3.1 Flash Lite 86/100, Macaron V1 Tall 53/100, MiMo-V2.5-Pro 47/100) 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, Gemini 3.1 Flash Lite, Macaron V1 Tall or MiMo-V2.5-Pro?

Macaron V1 Tall is cheaper at $0.247 input / $1.43 output per million tokens (median across 1 API provider). MiMo-V2.5-Pro costs $0.435 input / $0.87 output per million tokens (official Xiaomi API price); Gemini 3.1 Flash Lite costs $0.25 input / $1.50 output per million tokens (official Google API price). At a typical mix of three input tokens to one output token, that is $0.543 per million tokens for Macaron V1 Tall versus $0.544 for MiMo-V2.5-Pro (1× as much) and $0.563 for Gemini 3.1 Flash Lite (1× as much).

Which scores higher on benchmarks?

There is no independent benchmark that covers all three models yet. Gemini 3.1 Flash Lite has not been scored yet, Macaron V1 Tall has not been scored yet and MiMo-V2.5-Pro has not been scored yet.

Which is better for coding?

There are no published SWE-bench Verified results for Gemini 3.1 Flash Lite, Macaron V1 Tall and MiMo-V2.5-Pro 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?

Gemini 3.1 Flash Lite and MiMo-V2.5-Pro have the largest context windows (1,048,576 and 1,048,576 tokens), against 262,144 for Macaron V1 Tall. Maximum output per response: Gemini 3.1 Flash Lite up to 65,536, Macaron V1 Tall up to 32,768, MiMo-V2.5-Pro up to 131,072 tokens.

Which can read images, PDFs, audio or video?

Gemini 3.1 Flash Lite accepts text, images, PDFs, audio and video; Macaron V1 Tall accepts text and images; MiMo-V2.5-Pro accepts text. Gemini 3.1 Flash Lite handles the widest range of inputs.

Are any of these open source?

Macaron V1 Tall and MiMo-V2.5-Pro publishes its weights (MIT) and can be self-hosted; Gemini 3.1 Flash Lite is proprietary.

Which is newer?

Macaron V1 Tall is the newest, released Jul 21, 2026. Gemini 3.1 Flash Lite came out May 7, 2026; MiMo-V2.5-Pro came out Apr 22, 2026. Knowledge cutoff: Gemini 3.1 Flash Lite Jan 2025, MiMo-V2.5-Pro Dec 2024.

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.