ZMIME
Comparison · 3 models · Updated Nov 18, 2026

DeepSeek V4 Pro vs Gemini 3 Pro Preview vs GPT-5.2

Gemini 3 Pro Preview comes out ahead, 80 to 75 and 67 on our weighted score, though DeepSeek V4 Pro is 2.2× cheaper per token.

Price in rating: Excluded · Include price

  1. DeepSeek

    DeepSeek V4 Pro

    Released Apr 24, 2026

    67/100
    • ECI150.2
    • Price$1.60 / $3.33
    • Context1M
  2. Our pick

    Google

    Gemini 3 Pro Preview

    Released Nov 18, 2025

    80/100
    • ECI153.0
    • Price$2.00 / $12.00
    • Context1.05M
  3. OpenAI

    GPT-5.2

    Released Dec 11, 2025

    75/100
    • ECI154.5
    • Price$1.75 / $14.00
    • Context400K
01 — Verdict

Gemini 3 Pro Preview is our pick

Gemini 3 Pro Preview is the better all-round choice, scoring 80/100 against GPT-5.2 (75) and DeepSeek V4 Pro (67). It leads on inputs & features. The score weighs capability 67%, inputs & features 20%, context window 13%.

  • CapabilityGPT-5.2Capabilities Index (ECI): GPT-5.2 154.5 · Gemini 3 Pro Preview 153.0 · DeepSeek V4 Pro 150.2
  • Lowest priceDeepSeek V4 ProDeepSeek V4 Pro $2.03 · Gemini 3 Pro Preview $4.50 · GPT-5.2 $4.81 per 1M tokens (3:1 blend)
  • Longest contextGemini 3 Pro PreviewGemini 3 Pro Preview 1,048,576 · DeepSeek V4 Pro 1,000,000 · GPT-5.2 400,000 tokens
  • Widest inputsGemini 3 Pro PreviewDeepSeek V4 Pro: Text · Gemini 3 Pro Preview: Text, Images, PDFs, Audio, Video · GPT-5.2: Text, Images
  • Self-hostingDeepSeek V4 ProPublishes downloadable weights
How the score is built
MeasureWeightDeepSeek V4 ProGemini 3 Pro PreviewGPT-5.2
CapabilityCapabilities Index (ECI)67%778082
Inputs & features20%359070
Context window13%646550
Overall100%67/10080/10075/100
02 — Side by side

Every spec in one table

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

DeepSeek V4 Pro vs Gemini 3 Pro Preview vs GPT-5.2 specifications side by side
SpecificationDeepSeek V4 ProDeepSeekGemini 3 Pro PreviewGoogleGPT-5.2OpenAI
Capability
Capabilities Index (ECI)150.2153.0154.5 (best)
ECI rank#67 of 158#55 of 158#51 of 158 (best)
GPQA DiamondGraduate-level science questions90.9%92.6% (best)91.4%
FrontierMath Tiers 1–3Research-level mathematics45.3%—67.4% (best)
OTIS Mock AIME 2024–2025Competition mathematics96.7% (best)91.4%96.1%
SWE-bench VerifiedFixing real GitHub issues77.6% (best)72.9%73.8%
SimpleQA VerifiedShort factual questions47.0% (best)—37.1%
Price per million tokens
Input$1.60 (best)$2.00$1.75
Output$3.33 (best)$12.00$14.00
Cached input——$0.175
Blended (3:1)$2.03 (best)$4.50$4.81
Long-context rateSame rateSame rateSame rate
Price sourceMedian of 65 providersMedian of 6 providersOfficial OpenAI API
Limits
Context window1,000,000 tokens1,048,576 tokens (best)400,000 tokens
Max output384,000 tokens (best)65,536 tokens128,000 tokens
Inputs and features
TextYesYesYes
ImagesNoYesYes
PDFsNoYesNo
AudioNoYesNo
VideoNoYesNo
ReasoningYesYesYeslow · medium · high · xhigh
Tool callingYesYesYes
Structured outputNoNoYes
Availability
WeightsOpenProprietaryProprietary
API model ID——gpt-5.2
API providers67 (best)836
ReleasedApr 24, 2026Nov 18, 2025Dec 11, 2025
Knowledge cutoffMay 2025Jan 2025Aug 31, 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.

  • DeepSeek V4 Pro$22.66
  • Gemini 3 Pro Preview$44.00
  • GPT-5.2$45.50
04 — Questions

Which should you choose?

Which is better: DeepSeek V4 Pro, Gemini 3 Pro Preview or GPT-5.2?

Gemini 3 Pro Preview is the better all-round choice, scoring 80/100 against GPT-5.2 (75) and DeepSeek V4 Pro (67). It leads on inputs & features. The score weighs capability 67%, inputs & features 20%, context window 13%.

Which is cheaper, DeepSeek V4 Pro, Gemini 3 Pro Preview or GPT-5.2?

DeepSeek V4 Pro is cheaper at $1.60 input / $3.33 output per million tokens (median across 65 API providers). Gemini 3 Pro Preview costs $2.00 input / $12.00 output per million tokens (median across 6 API providers); GPT-5.2 costs $1.75 input / $14.00 output per million tokens (official OpenAI API price). At a typical mix of three input tokens to one output token, that is $2.03 per million tokens for DeepSeek V4 Pro versus $4.50 for Gemini 3 Pro Preview (2.2× as much) and $4.81 for GPT-5.2 (2.4× as much).

Which scores higher on benchmarks?

GPT-5.2 scores higher on the Capabilities Index (ECI): GPT-5.2 154.5 (#51 of 158), Gemini 3 Pro Preview 153.0 (#55 of 158) and DeepSeek V4 Pro 150.2 (#67 of 158). The confidence ranges of the top two overlap (152.9–156.3 vs 151.5–154.8), so treat the gap as small. On individual benchmarks: GPQA Diamond — Gemini 3 Pro Preview 92.6%, GPT-5.2 91.4%, DeepSeek V4 Pro 90.9%; OTIS Mock AIME 2024–2025 — DeepSeek V4 Pro 96.7%, GPT-5.2 96.1%, Gemini 3 Pro Preview 91.4%; SWE-bench Verified — DeepSeek V4 Pro 77.6%, GPT-5.2 73.8%, Gemini 3 Pro Preview 72.9%.

Which is better for coding?

DeepSeek V4 Pro resolves more real GitHub issues on SWE-bench Verified: DeepSeek V4 Pro 77.6%, GPT-5.2 73.8% and Gemini 3 Pro Preview 72.9%. All three support tool calling for agent workflows.

Which has the bigger context window?

Gemini 3 Pro Preview has the largest context window at 1,048,576 tokens, against 1,000,000 for DeepSeek V4 Pro and 400,000 for GPT-5.2. Maximum output per response: DeepSeek V4 Pro up to 384,000, Gemini 3 Pro Preview up to 65,536, GPT-5.2 up to 128,000 tokens.

Which can read images, PDFs, audio or video?

DeepSeek V4 Pro accepts text; Gemini 3 Pro Preview accepts text, images, PDFs, audio and video; GPT-5.2 accepts text and images. Gemini 3 Pro Preview handles the widest range of inputs.

Are any of these open source?

DeepSeek V4 Pro publishes its weights and can be self-hosted; Gemini 3 Pro Preview and GPT-5.2 is proprietary.

Which is newer?

DeepSeek V4 Pro is the newest, released Apr 24, 2026. GPT-5.2 came out Dec 11, 2025; Gemini 3 Pro Preview came out Nov 18, 2025. Knowledge cutoff: DeepSeek V4 Pro May 2025, Gemini 3 Pro Preview Jan 2025, GPT-5.2 Aug 31, 2025.

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.