ZMIME
Comparison · 2 models · Updated Nov 18, 2026

Gemini 3.1 Pro Preview vs Gemini 4 Carbon

Gemini 4 Carbon comes out ahead, 75 to 68 on our weighted score, and it is the cheaper option too.

Price in rating: Included · Exclude price

  1. Google

    Gemini 3.1 Pro Preview

    Released Feb 19, 2026

    68/100
    • ECI156.3
    • Price$2.00 / $12.00
    • Context1.05M
  2. Our pick

    Google

    Gemini 4 Carbon

    Released Nov 10, 2026New

    75/100
    • ECI169.8
    • Price$2.00 / $10.00
    • Context1.05M
  3. Add a model

    Make it a three-way comparison.

01 — Verdict

Gemini 4 Carbon is our pick

Gemini 4 Carbon is the better all-round choice, scoring 75/100 against Gemini 3.1 Pro Preview (68). It leads on capability and price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.

  • CapabilityGemini 4 CarbonCapabilities Index (ECI): Gemini 4 Carbon 169.8 · Gemini 3.1 Pro Preview 156.3
  • Lowest priceGemini 4 CarbonGemini 4 Carbon $4.00 · Gemini 3.1 Pro Preview $4.50 per 1M tokens (3:1 blend)
  • Longest contextAbout the sameGemini 3.1 Pro Preview 1,048,576 · Gemini 4 Carbon 1,048,576 tokens
  • Widest inputsSame inputsGemini 3.1 Pro Preview: Text, Images, PDFs, Audio, Video · Gemini 4 Carbon: Text, Images, PDFs, Audio, Video
  • Self-hostingNo confirmed open weightsBoth are available only through APIs
How the score is built
MeasureWeightGemini 3.1 Pro PreviewGemini 4 Carbon
CapabilityCapabilities Index (ECI)50%8396
Price25%1820
Inputs & features15%100100
Context window10%6565
Overall100%68/10075/100
02 — Side by side

Every spec in one table

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

Gemini 3.1 Pro Preview vs Gemini 4 Carbon specifications side by side
SpecificationGemini 3.1 Pro PreviewGoogleGemini 4 CarbonGoogle
Capability
Capabilities Index (ECI)156.3169.8 (best)
ECI rank#40 of 158#3 of 158 (best)
GPQA DiamondGraduate-level science questions94.4%96.5% (best)
FrontierMath Tiers 1–3Research-level mathematics59.7%90.5% (best)
OTIS Mock AIME 2024–2025Competition mathematics95.6%100% (best)
SWE-bench VerifiedFixing real GitHub issues75.6%—
SimpleQA VerifiedShort factual questions73.5%77.7% (best)
Price per million tokens
Input$2.00$2.00
Output$12.00$10.00 (best)
Cached input$0.20$0.10 (best)
Blended (3:1)$4.50$4.00 (best)
Long-context rateOver 200K: $4.00 / $18.00Same rate
Price sourceOfficial Google APIOfficial Google API
Limits
Context window1,048,576 tokens1,048,576 tokens
Max output65,536 tokens65,536 tokens
Inputs and features
TextYesYes
ImagesYesYes
PDFsYesYes
AudioYesYes
VideoYesYes
ReasoningYeslow · medium · highYeslow · medium · high
Tool callingYesYes
Structured outputYesYes
Availability
WeightsProprietaryProprietary
API model IDgemini-3.1-pro-previewgemini-4-carbon
API providers40 (best)18
ReleasedFeb 19, 2026Nov 10, 2026
Knowledge cutoffJan 2025Mar 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.

  • Gemini 3.1 Pro Preview$44.00
  • Gemini 4 Carbon$40.00
04 — Questions

Which should you choose?

Which is better: Gemini 3.1 Pro Preview or Gemini 4 Carbon?

Gemini 4 Carbon is the better all-round choice, scoring 75/100 against Gemini 3.1 Pro Preview (68). It leads on capability and price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.

Which is cheaper, Gemini 3.1 Pro Preview or Gemini 4 Carbon?

Gemini 4 Carbon is cheaper at $2.00 input / $10.00 output per million tokens (official Google API price). Gemini 3.1 Pro Preview costs $2.00 input / $12.00 output per million tokens (official Google API price). At a typical mix of three input tokens to one output token, that is $4.00 per million tokens for Gemini 4 Carbon versus $4.50 for Gemini 3.1 Pro Preview (1.1× as much).

Which scores higher on benchmarks?

Gemini 4 Carbon scores higher on the Capabilities Index (ECI): Gemini 4 Carbon 169.8 (#3 of 158) and Gemini 3.1 Pro Preview 156.3 (#40 of 158). Their confidence ranges do not overlap (166.6–173.3 vs 154.5–158.2), so the gap is a real one. On individual benchmarks: GPQA Diamond — Gemini 4 Carbon 96.5%, Gemini 3.1 Pro Preview 94.4%; FrontierMath Tiers 1–3 — Gemini 4 Carbon 90.5%, Gemini 3.1 Pro Preview 59.7%; OTIS Mock AIME 2024–2025 — Gemini 4 Carbon 100%, Gemini 3.1 Pro Preview 95.6%; SimpleQA Verified — Gemini 4 Carbon 77.7%, Gemini 3.1 Pro Preview 73.5%.

Which is better for coding?

There are no published SWE-bench Verified results for Gemini 4 Carbon yet, so there is no like-for-like coding score. On overall capability, Gemini 4 Carbon leads, which tends to carry over to coding, but test on your own codebase. Both support tool calling for agent workflows.

Which has the bigger context window?

Gemini 3.1 Pro Preview and Gemini 4 Carbon share the same 1,048,576-token context window. Maximum output per response: Gemini 3.1 Pro Preview up to 65,536, Gemini 4 Carbon up to 65,536 tokens.

Which can read images, PDFs, audio or video?

Gemini 3.1 Pro Preview accepts text, images, PDFs, audio and video; Gemini 4 Carbon accepts text, images, PDFs, audio and video. They handle the same number of input types.

Are any of these open source?

No. Gemini 3.1 Pro Preview and Gemini 4 Carbon are proprietary and only available through APIs and apps.

Which is newer?

Gemini 4 Carbon is the newest, released Nov 10, 2026. Gemini 3.1 Pro Preview came out Feb 19, 2026. Knowledge cutoff: Gemini 3.1 Pro Preview Jan 2025, Gemini 4 Carbon Mar 2026.

How do you decide the winner?

This rating uses capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, 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.