ZMIME
Comparison · 3 models · Updated Nov 19, 2026

SenseNova-Loop-9B vs Step 5 Preview vs GPT-6.2 Sol

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

Price in rating: Included · Exclude price

  1. SenseTime

    SenseNova-Loop-9B

    Released Nov 17, 2026New

    54/100
    • ECI—
    • Price—
    • Context524K
  2. StepFun

    Step 5 Preview

    Released Sep 16, 2026

    64/100
    • ECI—
    • Price$1.00 / $2.70
    • Context1M
  3. OpenAI

    GPT-6.2 Sol

    Released Nov 17, 2026New

    65/100
    • ECI—
    • Price$2.00 / $10.00
    • Context1.05M
01 — Verdict

Capability evidence is incomplete

There is not enough shared capability evidence to name an overall winner. The scores (GPT-6.2 Sol 65/100, Step 5 Preview 64/100, SenseNova-Loop-9B 54/100) compare the available specifications and price. The score weighs context window 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 priceStep 5 PreviewStep 5 Preview $1.43 · GPT-6.2 Sol $4.00 per 1M tokens (3:1 blend) · SenseNova-Loop-9B unpriced
  • Longest contextGPT-6.2 SolGPT-6.2 Sol 1,050,000 · Step 5 Preview 1,000,000 · SenseNova-Loop-9B 524,288 tokens
  • Widest inputsStep 5 Preview and GPT-6.2 SolSenseNova-Loop-9B: Text, Images · Step 5 Preview: Text, Images, Video · GPT-6.2 Sol: Text, Images, PDFs
  • Self-hostingSenseNova-Loop-9B and Step 5 PreviewPublishes downloadable weights
How the score is built
MeasureWeightSenseNova-Loop-9BStep 5 PreviewGPT-6.2 Sol
Context window100%546465
Specs & price only100%54/10064/10065/100

Left out because at least one model lacks the data: capability, price and inputs & features. 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.

SenseNova-Loop-9B vs Step 5 Preview vs GPT-6.2 Sol specifications side by side
SpecificationSenseNova-Loop-9BSenseTimeStep 5 PreviewStepFunGPT-6.2 SolOpenAI
TaskVision-language reasoning——
Published capabilitiesReuses middle layers to perform 56 layer executions with 32 physical decoder layers.StepFun's next-generation flagship base model for coding and professional knowledge work, with native text, image, and video input and a 1M-token context windowAn efficient reasoning model for everyday work, multistep agentic workflows, and coding.
Model size9B parameters——
AccessPublic weights including quantized variants for local inference——
SourcesModel weights——
Capability
Capabilities Index (ECI)———
ECI rank———
Price per million tokens
Input—$1.00 (best)$2.00
Output—$2.70 (best)$10.00
Cached input—$0.10$0.10
Blended (3:1)—$1.43 (best)$4.00
Long-context rate—Same rateOver 272K: $4.00 / $15.00
Price source—Official StepFun (Global) APIOfficial OpenAI API
Limits
Context window524,288 tokens1,000,000 tokens1,050,000 tokens (best)
Max output—1,000,000 tokens (best)128,000 tokens
Inputs and features
TextYesYesYes
ImagesYesYesYes
PDFsNoNoYes
AudioNoNoNo
VideoNoYesNo
ReasoningUnconfirmedYeslow · medium · highYeslow · medium · high · xhigh · max
Tool calling—YesYes
Structured output—YesYes
Availability
WeightsOpenOpenProprietary
API model ID—step-5-previewgpt-6.2-sol
API providers—2122 (best)
ReleasedNov 17, 2026Sep 16, 2026Nov 17, 2026
Knowledge cutoff——Jun 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.

  • SenseNova-Loop-9B—
  • Step 5 Preview$15.40
  • GPT-6.2 Sol$40.00
04 — Questions

Which should you choose?

Which is better: SenseNova-Loop-9B, Step 5 Preview or GPT-6.2 Sol?

There is not enough shared capability evidence to name an overall winner. The scores (GPT-6.2 Sol 65/100, Step 5 Preview 64/100, SenseNova-Loop-9B 54/100) compare the available specifications and price. The score weighs context window 100%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.

Which is cheaper, SenseNova-Loop-9B, Step 5 Preview or GPT-6.2 Sol?

Step 5 Preview is cheaper at $1.00 input / $2.70 output per million tokens (official StepFun (Global) API price). GPT-6.2 Sol costs $2.00 input / $10.00 output per million tokens (official OpenAI API price). At a typical mix of three input tokens to one output token, that is $1.43 per million tokens for Step 5 Preview versus $4.00 for GPT-6.2 Sol (2.8× as much). SenseNova-Loop-9B has no published per-token price.

Which scores higher on benchmarks?

There is no independent benchmark that covers all three models yet. SenseNova-Loop-9B has not been scored yet, Step 5 Preview has not been scored yet and GPT-6.2 Sol has not been scored yet.

Which is better for coding?

There are no published SWE-bench Verified results for SenseNova-Loop-9B, Step 5 Preview and GPT-6.2 Sol yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. Note that SenseNova-Loop-9B has no confirmed tool-calling support in this comparison.

Which has the bigger context window?

GPT-6.2 Sol has the largest context window at 1,050,000 tokens, against 1,000,000 for Step 5 Preview and 524,288 for SenseNova-Loop-9B. Maximum output per response: Step 5 Preview up to 1,000,000, GPT-6.2 Sol up to 128,000 tokens.

Which can read images, PDFs, audio or video?

SenseNova-Loop-9B accepts text and images; Step 5 Preview accepts text, images and video; GPT-6.2 Sol accepts text, images and PDFs. Step 5 Preview handles the widest range of inputs.

Are any of these open source?

SenseNova-Loop-9B and Step 5 Preview publishes its weights and can be self-hosted; GPT-6.2 Sol is proprietary.

Which is newer?

SenseNova-Loop-9B is the newest, released Nov 17, 2026. GPT-6.2 Sol came out Nov 17, 2026; Step 5 Preview came out Sep 16, 2026. Knowledge cutoff: GPT-6.2 Sol Jun 1, 2026.

How do you decide the winner?

This rating uses context window (100%, 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 19, 2026.