ZMIME
Comparison · 3 models · Updated Nov 18, 2026

Llama-3.2-90B-Vision-Instruct vs Gemma 3 12B IT vs GPT-4o mini

Too close to call on our weighted score (GPT-4o mini 33, Gemma 3 12B IT 33, Llama-3.2-90B-Vision-Instruct 29). The right pick depends on what you value most.

Price in rating: Included · Exclude price

  1. Meta

    Llama-3.2-90B-Vision-Instruct

    Released Sep 25, 2024

    29/100
    • ECI122.9
    • Price—
    • Context128K
  2. Google

    Gemma 3 12B IT

    Released Mar 12, 2025

    33/100
    • ECI—
    • Price$0.05 / $0.15
    • Context131K
  3. OpenAI

    GPT-4o mini

    Released Jul 18, 2024

    33/100
    • ECI123.8
    • Price$0.15 / $0.60
    • Context128K
01 — Verdict

Too close to call

It is close. Our weighted score puts them within a point (GPT-4o mini 33/100, Gemma 3 12B IT 33/100, Llama-3.2-90B-Vision-Instruct 29/100), so choose by what matters most for your work: Gemma 3 12B IT for raw 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 has no Capabilities Index score yet.

  • CapabilityGemma 3 12B ITShared benchmarks: Gemma 3 12B IT 28.1% · GPT-4o mini 22.3% · Llama-3.2-90B-Vision-Instruct 21.8%
  • Lowest priceGemma 3 12B ITGemma 3 12B IT $0.075 · GPT-4o mini $0.263 per 1M tokens (3:1 blend) · Llama-3.2-90B-Vision-Instruct unpriced
  • Longest contextGemma 3 12B ITGemma 3 12B IT 131,072 · Llama-3.2-90B-Vision-Instruct 128,000 · GPT-4o mini 128,000 tokens
  • Widest inputsGPT-4o miniLlama-3.2-90B-Vision-Instruct: Text, Images · Gemma 3 12B IT: Text, Images · GPT-4o mini: Text, Images, PDFs
  • Self-hostingLlama-3.2-90B-Vision-Instruct and Gemma 3 12B ITPublishes downloadable weights (Llama 3.2 Community License)
How the score is built
MeasureWeightLlama-3.2-90B-Vision-InstructGemma 3 12B ITGPT-4o mini
CapabilityShared benchmarks67%222822
Inputs & features20%505070
Context window13%323232
Overall100%29/10033/10033/100

Left out because at least one model lacks the data: price. 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.

Llama-3.2-90B-Vision-Instruct vs Gemma 3 12B IT vs GPT-4o mini specifications side by side
SpecificationLlama-3.2-90B-Vision-InstructMetaGemma 3 12B ITGoogleGPT-4o miniOpenAI
Capability
Capabilities Index (ECI)122.9—123.8 (best)
ECI rank#148 of 158—#147 of 158 (best)
GPQA DiamondGraduate-level science questions41.0% (best)39.5%37.7%
FrontierMath Tiers 1–3Research-level mathematics——0.7%
OTIS Mock AIME 2024–2025Competition mathematics2.6%16.7% (best)6.9%
SimpleQA VerifiedShort factual questions——8.3%
Price per million tokens
Input—$0.05 (best)$0.15
Output—$0.15 (best)$0.60
Cached input——$0.075
Blended (3:1)—$0.075 (best)$0.263
Long-context rate—Same rateSame rate
Price source—Median of 13 providersOfficial OpenAI API
Limits
Context window128,000 tokens131,072 tokens (best)128,000 tokens
Max output4,096 tokens131,072 tokens (best)16,384 tokens
Inputs and features
TextYesYesYes
ImagesYesYesYes
PDFsNoNoYes
AudioNoNoNo
VideoNoNoNo
ReasoningNoNoNo
Tool callingYesYesYes
Structured outputNoNoYes
Availability
WeightsOpenLlama 3.2 Community LicenseOpenProprietary
API model ID——gpt-4o-mini
API providers11427 (best)
ReleasedSep 25, 2024Mar 12, 2025Jul 18, 2024
Knowledge cutoffDec 2023Aug 2024Sep 2023
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.

  • Llama-3.2-90B-Vision-Instruct—
  • Gemma 3 12B IT$0.80
  • GPT-4o mini$2.70
04 — Questions

Which should you choose?

Which is better: Llama-3.2-90B-Vision-Instruct, Gemma 3 12B IT or GPT-4o mini?

It is close. Our weighted score puts them within a point (GPT-4o mini 33/100, Gemma 3 12B IT 33/100, Llama-3.2-90B-Vision-Instruct 29/100), so choose by what matters most for your work: Gemma 3 12B IT for raw 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 has no Capabilities Index score yet.

Which is cheaper, Llama-3.2-90B-Vision-Instruct, Gemma 3 12B IT or GPT-4o mini?

Gemma 3 12B IT is cheaper at $0.05 input / $0.15 output per million tokens (median across 13 API providers). GPT-4o mini costs $0.15 input / $0.60 output per million tokens (official OpenAI API price). At a typical mix of three input tokens to one output token, that is $0.075 per million tokens for Gemma 3 12B IT versus $0.263 for GPT-4o mini (3.5× as much). Llama-3.2-90B-Vision-Instruct has no published per-token price.

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%, GPT-4o mini 22.3% and Llama-3.2-90B-Vision-Instruct 21.8%. On individual benchmarks: GPQA Diamond — Llama-3.2-90B-Vision-Instruct 41.0%, Gemma 3 12B IT 39.5%, GPT-4o mini 37.7%; OTIS Mock AIME 2024–2025 — Gemma 3 12B IT 16.7%, GPT-4o mini 6.9%, Llama-3.2-90B-Vision-Instruct 2.6%.

Which is better for coding?

There are no published SWE-bench Verified results for Llama-3.2-90B-Vision-Instruct, Gemma 3 12B IT and GPT-4o mini 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 has the largest context window at 131,072 tokens, against 128,000 for Llama-3.2-90B-Vision-Instruct and 128,000 for GPT-4o mini. Maximum output per response: Llama-3.2-90B-Vision-Instruct up to 4,096, Gemma 3 12B IT up to 131,072, GPT-4o mini up to 16,384 tokens.

Which can read images, PDFs, audio or video?

Llama-3.2-90B-Vision-Instruct accepts text and images; Gemma 3 12B IT accepts text and images; GPT-4o mini accepts text, images and PDFs. GPT-4o mini handles the widest range of inputs.

Are any of these open source?

Llama-3.2-90B-Vision-Instruct and Gemma 3 12B IT publishes its weights (Llama 3.2 Community License) and can be self-hosted; GPT-4o mini is proprietary.

Which is newer?

Gemma 3 12B IT is the newest, released Mar 12, 2025. Llama-3.2-90B-Vision-Instruct came out Sep 25, 2024; GPT-4o mini came out Jul 18, 2024. Knowledge cutoff: Llama-3.2-90B-Vision-Instruct Dec 2023, Gemma 3 12B IT Aug 2024, GPT-4o mini Sep 2023.

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.