ZMIME
Comparison · 3 models · Updated Nov 18, 2026

Gemma 3 12B IT vs Gemma 3 4B IT vs Qwen2.5 7B Instruct

Gemma 3 12B IT comes out ahead, 33 to 25 and 22 on our weighted score, though Gemma 3 4B IT is 17% cheaper per token.

Price in rating: Excluded · Include price

  1. Our pick

    Google

    Gemma 3 12B IT

    Released Mar 12, 2025

    33/100
    • ECI—
    • Price$0.05 / $0.15
    • Context131K
  2. Google

    Gemma 3 4B IT

    Released Mar 12, 2025

    25/100
    • ECI—
    • Price$0.05 / $0.10
    • Context131K
  3. Alibaba (Qwen)

    Qwen2.5 7B Instruct

    Released Sep 19, 2024

    22/100
    • ECI—
    • Price$0.075 / $0.15
    • Context131K
01 — Verdict

Gemma 3 12B IT is our pick

Gemma 3 12B IT is the better all-round choice, scoring 33/100 against Gemma 3 4B IT (25) and Qwen2.5 7B Instruct (22). It leads on capability. The score weighs capability 67%, inputs & features 20%, context window 13%. Capability uses the average of shared benchmarks (GPQA Diamond and OTIS Mock AIME 2024–2025), because Gemma 3 12B IT, Gemma 3 4B IT and Qwen2.5 7B Instruct has no Capabilities Index score yet.

  • CapabilityGemma 3 12B ITShared benchmarks: Gemma 3 12B IT 28.1% · Qwen2.5 7B Instruct 19.0% · Gemma 3 4B IT 15.4%
  • Lowest priceGemma 3 4B ITGemma 3 4B IT $0.063 · Gemma 3 12B IT $0.075 · Qwen2.5 7B Instruct $0.094 per 1M tokens (3:1 blend)
  • Longest contextAbout the sameGemma 3 12B IT 131,072 · Gemma 3 4B IT 131,072 · Qwen2.5 7B Instruct 131,072 tokens
  • Widest inputsGemma 3 12B IT and Gemma 3 4B ITGemma 3 12B IT: Text, Images · Gemma 3 4B IT: Text, Images · Qwen2.5 7B Instruct: Text
  • Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
How the score is built
MeasureWeightGemma 3 12B ITGemma 3 4B ITQwen2.5 7B Instruct
CapabilityShared benchmarks67%281519
Inputs & features20%505025
Context window13%323232
Overall100%33/10025/10022/100
02 — Side by side

Every spec in one table

Highlighted cells lead their row. Dashes mean the data is not published.

Gemma 3 12B IT vs Gemma 3 4B IT vs Qwen2.5 7B Instruct specifications side by side
SpecificationGemma 3 12B ITGoogleGemma 3 4B ITGoogleQwen2.5 7B InstructAlibaba (Qwen)
Capability
Capabilities Index (ECI)———
ECI rank———
GPQA DiamondGraduate-level science questions39.5% (best)23.2%35.5%
OTIS Mock AIME 2024–2025Competition mathematics16.7% (best)7.5%2.5%
Price per million tokens
Input$0.05 (best)$0.05 (best)$0.075
Output$0.15$0.10 (best)$0.15
Cached input———
Blended (3:1)$0.075$0.063 (best)$0.094
Long-context rateSame rateSame rateSame rate
Price sourceMedian of 13 providersMedian of 9 providersMedian of 4 providers
Limits
Context window131,072 tokens131,072 tokens131,072 tokens
Max output131,072 tokens (best)131,072 tokens (best)8,192 tokens
Inputs and features
TextYesYesYes
ImagesYesYesNo
PDFsNoNoNo
AudioNoNoNo
VideoNoNoNo
ReasoningNoNoNo
Tool callingYesYesYes
Structured outputNoNoNo
Availability
WeightsOpenOpenOpenApache 2.0
API model ID———
API providers14 (best)105
ReleasedMar 12, 2025Mar 12, 2025Sep 19, 2024
Knowledge cutoffAug 2024Aug 2024—
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.

  • Gemma 3 12B IT$0.80
  • Gemma 3 4B IT$0.70
  • Qwen2.5 7B Instruct$1.05
04 — Questions

Which should you choose?

Which is better: Gemma 3 12B IT, Gemma 3 4B IT or Qwen2.5 7B Instruct?

Gemma 3 12B IT is the better all-round choice, scoring 33/100 against Gemma 3 4B IT (25) and Qwen2.5 7B Instruct (22). It leads on capability. The score weighs capability 67%, inputs & features 20%, context window 13%. Capability uses the average of shared benchmarks (GPQA Diamond and OTIS Mock AIME 2024–2025), because Gemma 3 12B IT, Gemma 3 4B IT and Qwen2.5 7B Instruct has no Capabilities Index score yet.

Which is cheaper, Gemma 3 12B IT, Gemma 3 4B IT or Qwen2.5 7B Instruct?

Gemma 3 4B IT is cheaper at $0.05 input / $0.10 output per million tokens (median across 9 API providers). Gemma 3 12B IT costs $0.05 input / $0.15 output per million tokens (median across 13 API providers); Qwen2.5 7B Instruct costs $0.075 input / $0.15 output per million tokens (median across 4 API providers). At a typical mix of three input tokens to one output token, that is $0.063 per million tokens for Gemma 3 4B IT versus $0.075 for Gemma 3 12B IT (1.2× as much) and $0.094 for Qwen2.5 7B Instruct (1.5× as much).

Which scores higher on benchmarks?

Not every model here has a Capabilities Index score, so we compare the average of shared benchmarks (GPQA Diamond and OTIS Mock AIME 2024–2025): Gemma 3 12B IT 28.1%, Qwen2.5 7B Instruct 19.0% and Gemma 3 4B IT 15.4%. On individual benchmarks: GPQA Diamond — Gemma 3 12B IT 39.5%, Qwen2.5 7B Instruct 35.5%, Gemma 3 4B IT 23.2%; OTIS Mock AIME 2024–2025 — Gemma 3 12B IT 16.7%, Gemma 3 4B IT 7.5%, Qwen2.5 7B Instruct 2.5%.

Which is better for coding?

There are no published SWE-bench Verified results for Gemma 3 12B IT, Gemma 3 4B IT and Qwen2.5 7B Instruct yet, so there is no like-for-like coding score. On overall capability, Gemma 3 12B IT leads, which tends to carry over to coding, but test on your own codebase. All three support tool calling for agent workflows.

Which has the bigger context window?

Gemma 3 12B IT, Gemma 3 4B IT and Qwen2.5 7B Instruct share the same 131,072-token context window. Maximum output per response: Gemma 3 12B IT up to 131,072, Gemma 3 4B IT up to 131,072, Qwen2.5 7B Instruct up to 8,192 tokens.

Which can read images, PDFs, audio or video?

Gemma 3 12B IT accepts text and images; Gemma 3 4B IT accepts text and images; Qwen2.5 7B Instruct accepts text. Gemma 3 12B 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?

Gemma 3 12B IT is the newest, released Mar 12, 2025. Gemma 3 4B IT came out Mar 12, 2025; Qwen2.5 7B Instruct came out Sep 19, 2024. Knowledge cutoff: Gemma 3 12B IT Aug 2024, Gemma 3 4B IT Aug 2024.

How do you decide the winner?

This rating uses capability (67%, independent benchmark results); inputs & features (20%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (13%, 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.