ZMIME
Comparison · 3 models · Updated Oct 4, 2026

MiniMax-M2.7 vs DeepSeek V3.2 vs GPT-5.4 nano

GPT-5.4 nano comes out ahead, 68 to 64 and 61 on our weighted score, though DeepSeek V3.2 is 26% cheaper per token.

  1. MiniMax

    MiniMax-M2.7

    Released Mar 18, 2026

    61/100
    • ECI145.9
    • Price$0.30 / $1.20
    • Context205K
    Full specs Remove MiniMax-M2.7

  2. DeepSeek

    DeepSeek V3.2

    Released Dec 1, 2025

    64/100
    • ECI146.3
    • Price$0.296 / $0.48
    • Context128K
    Full specs Remove DeepSeek V3.2
  3. Our pick

    OpenAI

    GPT-5.4 nano

    Released Mar 17, 2026

    68/100
    • ECI145.8
    • Price$0.20 / $1.25
    • Context400K
    Full specs Remove GPT-5.4 nano
01 — Verdict

GPT-5.4 nano is our pick

GPT-5.4 nano is the better all-round choice, scoring 68/100 against DeepSeek V3.2 (64) and MiniMax-M2.7 (61). It leads on inputs & features and context window. DeepSeek V3.2 wins on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.

  • CapabilityDeepSeek V3.2Capabilities Index (ECI): DeepSeek V3.2 146.3 · MiniMax-M2.7 145.9 · GPT-5.4 nano 145.8
  • Lowest priceDeepSeek V3.2DeepSeek V3.2 $0.342 · GPT-5.4 nano $0.463 · MiniMax-M2.7 $0.525 per 1M tokens (3:1 blend)
  • Longest contextGPT-5.4 nanoGPT-5.4 nano 400,000 · MiniMax-M2.7 204,800 · DeepSeek V3.2 128,000 tokens
  • Widest inputsGPT-5.4 nanoMiniMax-M2.7: Text · DeepSeek V3.2: Text · GPT-5.4 nano: Text, Images
  • Self-hostingMiniMax-M2.7 and DeepSeek V3.2Publishes downloadable weights (MIT License)
How the score is built
MeasureWeightMiniMax-M2.7DeepSeek V3.2GPT-5.4 nano
CapabilityCapabilities Index (ECI)50%737373
Price25%637266
Inputs & features15%354570
Context window10%322444
Overall100%61/10064/10068/100
02 — Side by side

Every spec in one table

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

MiniMax-M2.7 vs DeepSeek V3.2 vs GPT-5.4 nano specifications side by side
SpecificationMiniMax-M2.7MiniMaxDeepSeek V3.2DeepSeekGPT-5.4 nanoOpenAI
Capability
Capabilities Index (ECI)145.9146.3 (best)145.8
ECI rank#73 of 148#69 of 148 (best)#75 of 148
GPQA DiamondGraduate-level science questions—83.4% (best)78.5%
FrontierMath Tiers 1–3Research-level mathematics——44.9%
OTIS Mock AIME 2024–2025Competition mathematics—87.8% (best)87.8%
SimpleQA VerifiedShort factual questions——11.7%
Price per million tokens
Input$0.30$0.296$0.20 (best)
Output$1.20$0.48 (best)$1.25
Cached input$0.06—$0.02 (best)
Blended (3:1)$0.525$0.342 (best)$0.463
Long-context rateSame rateSame rateSame rate
Price sourceOfficial MiniMax (minimax.io) APIMedian of 15 providersOfficial OpenAI API
Limits
Context window204,800 tokens128,000 tokens400,000 tokens (best)
Max output131,072 tokens (best)64,000 tokens128,000 tokens
Inputs and features
TextYesYesYes
ImagesNoNoYes
PDFsNoNoNo
AudioNoNoNo
VideoNoNoNo
ReasoningYesYesYeslow · medium · high · xhigh
Tool callingYesYesYes
Structured outputNoYesYes
Availability
WeightsOpenOpenMIT LicenseProprietary
API model IDMiniMax-M2.7—gpt-5.4-nano
API providers29 (best)1526
ReleasedMar 18, 2026Dec 1, 2025Mar 17, 2026
Knowledge cutoff—Jul 2024Aug 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.

  • MiniMax-M2.7$5.40
  • DeepSeek V3.2$3.92
  • GPT-5.4 nano$4.50
04 — Questions

Which should you choose?

Which is better: MiniMax-M2.7, DeepSeek V3.2 or GPT-5.4 nano?

GPT-5.4 nano is the better all-round choice, scoring 68/100 against DeepSeek V3.2 (64) and MiniMax-M2.7 (61). It leads on inputs & features and context window. DeepSeek V3.2 wins on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.

Which is cheaper, MiniMax-M2.7, DeepSeek V3.2 or GPT-5.4 nano?

DeepSeek V3.2 is cheaper at $0.296 input / $0.48 output per million tokens (median across 15 API providers). GPT-5.4 nano costs $0.20 input / $1.25 output per million tokens (official OpenAI API price); MiniMax-M2.7 costs $0.30 input / $1.20 output per million tokens (official MiniMax (minimax.io) API price). At a typical mix of three input tokens to one output token, that is $0.342 per million tokens for DeepSeek V3.2 versus $0.463 for GPT-5.4 nano (1.4× as much) and $0.525 for MiniMax-M2.7 (1.5× as much).

Which scores higher on benchmarks?

DeepSeek V3.2 scores higher on the Capabilities Index (ECI): DeepSeek V3.2 146.3 (#69 of 148), MiniMax-M2.7 145.9 (#73 of 148) and GPT-5.4 nano 145.8 (#75 of 148). The confidence ranges of the top two overlap (144.4–147.5 vs 138.2–148.0), so treat the gap as small.

Which is better for coding?

There are no published SWE-bench Verified results for MiniMax-M2.7, DeepSeek V3.2 and GPT-5.4 nano yet, so there is no like-for-like coding score. On overall capability, DeepSeek V3.2 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?

GPT-5.4 nano has the largest context window at 400,000 tokens, against 204,800 for MiniMax-M2.7 and 128,000 for DeepSeek V3.2. Maximum output per response: MiniMax-M2.7 up to 131,072, DeepSeek V3.2 up to 64,000, GPT-5.4 nano up to 128,000 tokens.

Which can read images, PDFs, audio or video?

MiniMax-M2.7 accepts text; DeepSeek V3.2 accepts text; GPT-5.4 nano accepts text and images. GPT-5.4 nano handles the widest range of inputs.

Are any of these open source?

MiniMax-M2.7 and DeepSeek V3.2 publishes its weights (MIT License) and can be self-hosted; GPT-5.4 nano is proprietary.

Which is newer?

MiniMax-M2.7 is the newest, released Mar 18, 2026. GPT-5.4 nano came out Mar 17, 2026; DeepSeek V3.2 came out Dec 1, 2025. Knowledge cutoff: DeepSeek V3.2 Jul 2024, GPT-5.4 nano Aug 31, 2025.

How do you decide the winner?

Each model gets a 0–100 score on 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 Oct 4, 2026.

05 — Keep comparing

Related comparisons