ZMIME
Comparison · 3 models · Updated Nov 18, 2026

Step 3.7 Flash vs Mercury Edit 2 vs Perceptron Mk1

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. StepFun

    Step 3.7 Flash

    Released May 29, 2026

    65/100
    • ECI—
    • Price$0.20 / $1.15
    • Context256K
  2. Inception

    Mercury Edit 2

    Released Mar 30, 2026

    36/100
    • ECI—
    • Price$0.25 / $0.75
    • Context32K
  3. Perceptron

    Perceptron Mk1

    Released May 12, 2026

    47/100
    • ECI—
    • Price$0.15 / $1.50
    • Context33K
01 — Verdict

Capability evidence is incomplete

There is not enough shared capability evidence to name an overall winner. The scores (Step 3.7 Flash 65/100, Perceptron Mk1 47/100, Mercury Edit 2 36/100) 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 priceMercury Edit 2Mercury Edit 2 $0.375 · Step 3.7 Flash $0.438 · Perceptron Mk1 $0.487 per 1M tokens (3:1 blend)
  • Longest contextStep 3.7 FlashStep 3.7 Flash 256,000 · Perceptron Mk1 32,768 · Mercury Edit 2 32,000 tokens
  • Widest inputsStep 3.7 Flash and Perceptron Mk1Step 3.7 Flash: Text, Images, Video · Mercury Edit 2: Text · Perceptron Mk1: Text, Images, Video
  • Self-hostingStep 3.7 FlashPublishes downloadable weights
How the score is built
MeasureWeightStep 3.7 FlashMercury Edit 2Perceptron Mk1
Price50%646762
Inputs & features30%80045
Context window20%431011
Specs & price only100%65/10036/10047/100

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.

Step 3.7 Flash vs Mercury Edit 2 vs Perceptron Mk1 specifications side by side
SpecificationStep 3.7 FlashStepFunMercury Edit 2InceptionPerceptron Mk1Perceptron
Capability
Capabilities Index (ECI)———
ECI rank———
Price per million tokens
Input$0.20$0.25$0.15 (best)
Output$1.15$0.75 (best)$1.50
Cached input$0.04$0.025 (best)—
Blended (3:1)$0.438$0.375 (best)$0.487
Long-context rateSame rateSame rateSame rate
Price sourceOfficial StepFun (Global) APIOfficial Inception APIMedian of 3 providers
Limits
Context window256,000 tokens (best)32,000 tokens32,768 tokens
Max output256,000 tokens (best)8,192 tokens8,192 tokens
Inputs and features
TextYesYesYes
ImagesYesNoYes
PDFsNoNoNo
AudioNoNoNo
VideoYesNoYes
ReasoningYeslow · medium · highNoYes
Tool callingYesNoNo
Structured outputYesNoNo
Availability
WeightsOpenProprietaryProprietary
API model IDstep-3.7-flashmercury-edit-2—
API providers19 (best)13
ReleasedMay 29, 2026Mar 30, 2026May 12, 2026
Knowledge cutoffMar 1, 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.

  • Step 3.7 Flash$4.30
  • Mercury Edit 2$4.00
  • Perceptron Mk1$4.50
04 — Questions

Which should you choose?

Which is better: Step 3.7 Flash, Mercury Edit 2 or Perceptron Mk1?

There is not enough shared capability evidence to name an overall winner. The scores (Step 3.7 Flash 65/100, Perceptron Mk1 47/100, Mercury Edit 2 36/100) 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, Step 3.7 Flash, Mercury Edit 2 or Perceptron Mk1?

Mercury Edit 2 is cheaper at $0.25 input / $0.75 output per million tokens (official Inception API price). Step 3.7 Flash costs $0.20 input / $1.15 output per million tokens (official StepFun (Global) API price); Perceptron Mk1 costs $0.15 input / $1.50 output per million tokens (median across 3 API providers). At a typical mix of three input tokens to one output token, that is $0.375 per million tokens for Mercury Edit 2 versus $0.438 for Step 3.7 Flash (1.2× as much) and $0.487 for Perceptron Mk1 (1.3× as much).

Which scores higher on benchmarks?

There is no independent benchmark that covers all three models yet. Step 3.7 Flash has not been scored yet, Mercury Edit 2 has not been scored yet and Perceptron Mk1 has not been scored yet.

Which is better for coding?

There are no published SWE-bench Verified results for Step 3.7 Flash, Mercury Edit 2 and Perceptron Mk1 yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. Note that Mercury Edit 2 and Perceptron Mk1 does not support tool calling, which most coding agents need.

Which has the bigger context window?

Step 3.7 Flash has the largest context window at 256,000 tokens, against 32,768 for Perceptron Mk1 and 32,000 for Mercury Edit 2. Maximum output per response: Step 3.7 Flash up to 256,000, Mercury Edit 2 up to 8,192, Perceptron Mk1 up to 8,192 tokens.

Which can read images, PDFs, audio or video?

Step 3.7 Flash accepts text, images and video; Mercury Edit 2 accepts text; Perceptron Mk1 accepts text, images and video. Step 3.7 Flash handles the widest range of inputs.

Are any of these open source?

Step 3.7 Flash publishes its weights and can be self-hosted; Mercury Edit 2 and Perceptron Mk1 is proprietary.

Which is newer?

Step 3.7 Flash is the newest, released May 29, 2026. Perceptron Mk1 came out May 12, 2026; Mercury Edit 2 came out Mar 30, 2026. Knowledge cutoff: Step 3.7 Flash Mar 1, 2026.

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.