ZMIME
Comparison · 3 models · Updated Nov 18, 2026

Aion-3.0-Mini vs Kimi K2.5 vs Qwen3.5 27B

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. Aion Labs

    Aion-3.0-Mini

    Released May 14, 2026

    42/100
    • ECI—
    • Price$0.70 / $1.40
    • Context131K
  2. Moonshot AI

    Kimi K2.5

    Released Jan 27, 2026

    52/100
    • ECI148.8
    • Price$0.60 / $3.00
    • Context262K
  3. Alibaba (Qwen)

    Qwen3.5 27B

    Released Feb 23, 2026

    58/100
    • ECI—
    • Price$0.30 / $2.40
    • Context262K
01 — Verdict

Capability evidence is incomplete

There is not enough shared capability evidence to name an overall winner. The scores (Qwen3.5 27B 58/100, Kimi K2.5 52/100, Aion-3.0-Mini 42/100) 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 priceQwen3.5 27BQwen3.5 27B $0.825 · Aion-3.0-Mini $0.875 · Kimi K2.5 $1.20 per 1M tokens (3:1 blend)
  • Longest contextKimi K2.5 and Qwen3.5 27BKimi K2.5 262,144 · Qwen3.5 27B 262,144 · Aion-3.0-Mini 131,072 tokens
  • Widest inputsKimi K2.5 and Qwen3.5 27BAion-3.0-Mini: Text · Kimi K2.5: Text, Images, Video · Qwen3.5 27B: Text, Images, Video
  • Self-hostingKimi K2.5 and Qwen3.5 27BPublishes downloadable weights
How the score is built
MeasureWeightAion-3.0-MiniKimi K2.5Qwen3.5 27B
Price50%504452
Inputs & features30%357080
Context window20%324343
Specs & price only100%42/10052/10058/100

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.

Aion-3.0-Mini vs Kimi K2.5 vs Qwen3.5 27B specifications side by side
SpecificationAion-3.0-MiniAion LabsKimi K2.5Moonshot AIQwen3.5 27BAlibaba (Qwen)
Capability
Capabilities Index (ECI)—148.8—
ECI rank—#73 of 158—
GPQA DiamondGraduate-level science questions—87.6%—
OTIS Mock AIME 2024–2025Competition mathematics—92.2%—
SWE-bench VerifiedFixing real GitHub issues—73.8%—
SimpleQA VerifiedShort factual questions—34.3%—
Price per million tokens
Input$0.70$0.60$0.30 (best)
Output$1.40 (best)$3.00$2.40
Cached input———
Blended (3:1)$0.875$1.20$0.825 (best)
Long-context rateSame rateSame rateSame rate
Price sourceMedian of 4 providersMedian of 36 providersOfficial Alibaba API
Limits
Context window131,072 tokens262,144 tokens (best)262,144 tokens (best)
Max output32,768 tokens262,144 tokens (best)65,536 tokens
Inputs and features
TextYesYesYes
ImagesNoYesYes
PDFsNoNoNo
AudioNoNoNo
VideoNoYesYes
ReasoningYesYesYes
Tool callingYesYesYes
Structured outputNoNoYes
Availability
WeightsProprietaryOpenOpen
API model ID——qwen3.5-27b
API providers436 (best)22
ReleasedMay 14, 2026Jan 27, 2026Feb 23, 2026
Knowledge cutoff—Jan 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.

  • Aion-3.0-Mini$9.80
  • Kimi K2.5$12.00
  • Qwen3.5 27B$7.80
04 — Questions

Which should you choose?

Which is better: Aion-3.0-Mini, Kimi K2.5 or Qwen3.5 27B?

There is not enough shared capability evidence to name an overall winner. The scores (Qwen3.5 27B 58/100, Kimi K2.5 52/100, Aion-3.0-Mini 42/100) 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, Aion-3.0-Mini, Kimi K2.5 or Qwen3.5 27B?

Qwen3.5 27B is cheaper at $0.30 input / $2.40 output per million tokens (official Alibaba API price). Aion-3.0-Mini costs $0.70 input / $1.40 output per million tokens (median across 4 API providers); Kimi K2.5 costs $0.60 input / $3.00 output per million tokens (median across 36 API providers). At a typical mix of three input tokens to one output token, that is $0.825 per million tokens for Qwen3.5 27B versus $0.875 for Aion-3.0-Mini (1.1× as much) and $1.20 for Kimi K2.5 (1.5× as much).

Which scores higher on benchmarks?

There is no independent benchmark that covers all three models yet. Aion-3.0-Mini has not been scored yet, Kimi K2.5 has an ECI of 148.8 and Qwen3.5 27B has not been scored yet.

Which is better for coding?

There are no published SWE-bench Verified results for Aion-3.0-Mini and Qwen3.5 27B 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?

Kimi K2.5 and Qwen3.5 27B have the largest context windows (262,144 and 262,144 tokens), against 131,072 for Aion-3.0-Mini. Maximum output per response: Aion-3.0-Mini up to 32,768, Kimi K2.5 up to 262,144, Qwen3.5 27B up to 65,536 tokens.

Which can read images, PDFs, audio or video?

Aion-3.0-Mini accepts text; Kimi K2.5 accepts text, images and video; Qwen3.5 27B accepts text, images and video. Kimi K2.5 handles the widest range of inputs.

Are any of these open source?

Kimi K2.5 and Qwen3.5 27B publishes its weights and can be self-hosted; Aion-3.0-Mini is proprietary.

Which is newer?

Aion-3.0-Mini is the newest, released May 14, 2026. Qwen3.5 27B came out Feb 23, 2026; Kimi K2.5 came out Jan 27, 2026. Knowledge cutoff: Kimi K2.5 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.