ZMIME
Comparison · 3 models · Updated Nov 18, 2026

Hy-MT2-1.8B vs Gemma 4 31B IT vs DiffusionGemma 26B A4B IT

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. Tencent

    Hy-MT2-1.8B

    Released May 21, 2026

    49/100
    • ECI—
    • Price$0.044 / $0.177
    • Context8K
  2. Google

    Gemma 4 31B IT

    Released Apr 2, 2026

    69/100
    • ECI144.1
    • Price$0.142 / $0.40
    • Context262K
  3. Google

    DiffusionGemma 26B A4B IT

    Released Jun 9, 2026

    76/100
    • ECI—
    • Price$0.05 / $0.15
    • Context262K
01 — Verdict

Capability evidence is incomplete

There is not enough shared capability evidence to name an overall winner. The scores (DiffusionGemma 26B A4B IT 76/100, Gemma 4 31B IT 69/100, Hy-MT2-1.8B 49/100) 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 priceDiffusionGemma 26B A4B ITDiffusionGemma 26B A4B IT $0.075 · Hy-MT2-1.8B $0.077 · Gemma 4 31B IT $0.206 per 1M tokens (3:1 blend)
  • Longest contextGemma 4 31B IT and DiffusionGemma 26B A4B ITGemma 4 31B IT 262,144 · DiffusionGemma 26B A4B IT 262,144 · Hy-MT2-1.8B 8,192 tokens
  • Widest inputsGemma 4 31B IT and DiffusionGemma 26B A4B ITHy-MT2-1.8B: Text · Gemma 4 31B IT: Text, Images · DiffusionGemma 26B A4B IT: Text, Images
  • Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
How the score is built
MeasureWeightHy-MT2-1.8BGemma 4 31B ITDiffusionGemma 26B A4B IT
Price50%987999
Inputs & features30%07060
Context window20%04343
Specs & price only100%49/10069/10076/100

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.

Hy-MT2-1.8B vs Gemma 4 31B IT vs DiffusionGemma 26B A4B IT specifications side by side
SpecificationHy-MT2-1.8BTencentGemma 4 31B ITGoogleDiffusionGemma 26B A4B ITGoogle
Capability
Capabilities Index (ECI)—144.1—
ECI rank—#94 of 158—
GPQA DiamondGraduate-level science questions—75.8%—
OTIS Mock AIME 2024–2025Competition mathematics—73.3%—
SimpleQA VerifiedShort factual questions—10.4%—
Price per million tokens
Input$0.044 (best)$0.142$0.05
Output$0.177$0.40$0.15 (best)
Cached input———
Blended (3:1)$0.077$0.206$0.075 (best)
Long-context rateSame rateSame rateSame rate
Price sourceMedian of 3 providersMedian of 42 providersMedian of 1 providers
Limits
Context window8,192 tokens262,144 tokens (best)262,144 tokens (best)
Max output4,096 tokens32,768 tokens (best)32,768 tokens (best)
Inputs and features
TextYesYesYes
ImagesNoYesYes
PDFsNoNoNo
AudioNoNoNo
VideoNoNoNo
ReasoningNoYesYes
Tool callingNoYesYes
Structured outputNoYesNo
Availability
WeightsOpenOpenOpenApache 2.0
API model ID—gemma-4-31b-it—
API providers352 (best)3
ReleasedMay 21, 2026Apr 2, 2026Jun 9, 2026
Knowledge cutoff—Jan 2025Jan 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.

  • Hy-MT2-1.8B$0.794
  • Gemma 4 31B IT$2.22
  • DiffusionGemma 26B A4B IT$0.80
04 — Questions

Which should you choose?

Which is better: Hy-MT2-1.8B, Gemma 4 31B IT or DiffusionGemma 26B A4B IT?

There is not enough shared capability evidence to name an overall winner. The scores (DiffusionGemma 26B A4B IT 76/100, Gemma 4 31B IT 69/100, Hy-MT2-1.8B 49/100) 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, Hy-MT2-1.8B, Gemma 4 31B IT or DiffusionGemma 26B A4B IT?

DiffusionGemma 26B A4B IT is cheaper at $0.05 input / $0.15 output per million tokens (median across 1 API provider). Hy-MT2-1.8B costs $0.044 input / $0.177 output per million tokens (median across 3 API providers); Gemma 4 31B IT costs $0.142 input / $0.40 output per million tokens (median across 42 API providers; free on Google). At a typical mix of three input tokens to one output token, that is $0.075 per million tokens for DiffusionGemma 26B A4B IT versus $0.077 for Hy-MT2-1.8B (1× as much) and $0.206 for Gemma 4 31B IT (2.8× as much).

Which scores higher on benchmarks?

There is no independent benchmark that covers all three models yet. Hy-MT2-1.8B has not been scored yet, Gemma 4 31B IT has an ECI of 144.1 and DiffusionGemma 26B A4B IT has not been scored yet.

Which is better for coding?

There are no published SWE-bench Verified results for Hy-MT2-1.8B, Gemma 4 31B IT and DiffusionGemma 26B A4B IT yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. Note that Hy-MT2-1.8B does not support tool calling, which most coding agents need.

Which has the bigger context window?

Gemma 4 31B IT and DiffusionGemma 26B A4B IT have the largest context windows (262,144 and 262,144 tokens), against 8,192 for Hy-MT2-1.8B. Maximum output per response: Hy-MT2-1.8B up to 4,096, Gemma 4 31B IT up to 32,768, DiffusionGemma 26B A4B IT up to 32,768 tokens.

Which can read images, PDFs, audio or video?

Hy-MT2-1.8B accepts text; Gemma 4 31B IT accepts text and images; DiffusionGemma 26B A4B IT accepts text and images. Gemma 4 31B IT handles the widest range of inputs.

Are any of these open source?

Yes, all three publish their weights (Apache 2.0), so you can self-host them.

Which is newer?

DiffusionGemma 26B A4B IT is the newest, released Jun 9, 2026. Hy-MT2-1.8B came out May 21, 2026; Gemma 4 31B IT came out Apr 2, 2026. Knowledge cutoff: Gemma 4 31B IT Jan 2025, DiffusionGemma 26B A4B IT 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.