ZMIME
Comparison · 3 models · Updated Nov 18, 2026

Inkling Small vs Qwen3.8 27B vs Step 5 Preview

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. Thinking Machines

    Inkling Small

    Released Jul 30, 2026

    60/100
    • ECI150.8
    • Price$0.58 / $1.44
    • Context1.05M
  2. Alibaba (Qwen)

    Qwen3.8 27B

    Released Aug 14, 2026

    54/100
    • ECI152.7
    • Price$0.40 / $2.42
    • Context262K
  3. StepFun

    Step 5 Preview

    Released Sep 16, 2026

    57/100
    • ECI—
    • Price$1.00 / $2.70
    • Context1M
01 — Verdict

Capability evidence is incomplete

There is not enough shared capability evidence to name an overall winner. The scores (Inkling Small 60/100, Step 5 Preview 57/100, Qwen3.8 27B 54/100) 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 priceInkling SmallInkling Small $0.795 · Qwen3.8 27B $0.905 · Step 5 Preview $1.43 per 1M tokens (3:1 blend)
  • Longest contextInkling SmallInkling Small 1,048,576 · Step 5 Preview 1,000,000 · Qwen3.8 27B 262,144 tokens
  • Widest inputsSame inputsInkling Small: Text, Images, Audio · Qwen3.8 27B: Text, Images, Video · Step 5 Preview: Text, Images, Video
  • Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
How the score is built
MeasureWeightInkling SmallQwen3.8 27BStep 5 Preview
Price50%525041
Inputs & features30%707080
Context window20%654364
Specs & price only100%60/10054/10057/100

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.

Inkling Small vs Qwen3.8 27B vs Step 5 Preview specifications side by side
SpecificationInkling SmallThinking MachinesQwen3.8 27BAlibaba (Qwen)Step 5 PreviewStepFun
Capability
Capabilities Index (ECI)150.8152.7 (best)—
ECI rank#65 of 158#57 of 158 (best)—
GPQA DiamondGraduate-level science questions88.5%——
FrontierMath Tiers 1–3Research-level mathematics46.3%——
OTIS Mock AIME 2024–2025Competition mathematics90.0%——
SimpleQA VerifiedShort factual questions19.1%——
Price per million tokens
Input$0.58$0.40 (best)$1.00
Output$1.44 (best)$2.42$2.70
Cached input$0.116—$0.10 (best)
Blended (3:1)$0.795 (best)$0.905$1.43
Long-context rateSame rateSame rateSame rate
Price sourceOfficial Thinking Machines APIMedian of 67 providersOfficial StepFun (Global) API
Limits
Context window1,048,576 tokens (best)262,144 tokens1,000,000 tokens
Max output1,048,576 tokens (best)32,768 tokens1,000,000 tokens
Inputs and features
TextYesYesYes
ImagesYesYesYes
PDFsNoNoNo
AudioYesNoNo
VideoNoYesYes
ReasoningYeslow · medium · high · xhigh · maxYesYeslow · medium · high
Tool callingYesYesYes
Structured outputNoNoYes
Availability
WeightsOpenApache-2.0OpenOpen
API model IDthinkingmachines/Inkling-Small:peft:262144—step-5-preview
API providers1272 (best)21
ReleasedJul 30, 2026Aug 14, 2026Sep 16, 2026
Knowledge cutoff———
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.

  • Inkling Small$8.68
  • Qwen3.8 27B$8.84
  • Step 5 Preview$15.40
04 — Questions

Which should you choose?

Which is better: Inkling Small, Qwen3.8 27B or Step 5 Preview?

There is not enough shared capability evidence to name an overall winner. The scores (Inkling Small 60/100, Step 5 Preview 57/100, Qwen3.8 27B 54/100) 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, Inkling Small, Qwen3.8 27B or Step 5 Preview?

Inkling Small is cheaper at $0.58 input / $1.44 output per million tokens (official Thinking Machines API price). Qwen3.8 27B costs $0.40 input / $2.42 output per million tokens (median across 67 API providers); Step 5 Preview costs $1.00 input / $2.70 output per million tokens (official StepFun (Global) API price). At a typical mix of three input tokens to one output token, that is $0.795 per million tokens for Inkling Small versus $0.905 for Qwen3.8 27B (1.1× as much) and $1.43 for Step 5 Preview (1.8× as much).

Which scores higher on benchmarks?

There is no independent benchmark that covers all three models yet. Inkling Small has an ECI of 150.8, Qwen3.8 27B has an ECI of 152.7 and Step 5 Preview has not been scored yet.

Which is better for coding?

There are no published SWE-bench Verified results for Inkling Small, Qwen3.8 27B and Step 5 Preview 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?

Inkling Small has the largest context window at 1,048,576 tokens, against 1,000,000 for Step 5 Preview and 262,144 for Qwen3.8 27B. Maximum output per response: Inkling Small up to 1,048,576, Qwen3.8 27B up to 32,768, Step 5 Preview up to 1,000,000 tokens.

Which can read images, PDFs, audio or video?

Inkling Small accepts text, images and audio; Qwen3.8 27B accepts text, images and video; Step 5 Preview accepts text, images and video. They handle the same number of input types.

Are any of these open source?

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

Which is newer?

Step 5 Preview is the newest, released Sep 16, 2026. Qwen3.8 27B came out Aug 14, 2026; Inkling Small came out Jul 30, 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.