ZMIME
Comparison · 3 models · Updated Nov 18, 2026

DeepSeek V3.2 vs GPT-5.4 nano vs GPT OSS 120B

GPT-5.4 nano comes out ahead, 70 to 60 and 57 on our weighted score, though GPT OSS 120B is 43% cheaper per token.

Price in rating: Excluded · Include price

  1. DeepSeek

    DeepSeek V3.2

    Released Dec 1, 2025

    60/100
    • ECI146.3
    • Price$0.28 / $0.42
    • Context128K
  2. Our pick

    OpenAI

    GPT-5.4 nano

    Released Mar 17, 2026

    70/100
    • ECI146.3
    • Price$0.20 / $1.25
    • Context400K
  3. OpenAI

    GPT OSS 120B

    Released Aug 5, 2025

    57/100
    • ECI140.2
    • Price$0.15 / $0.60
    • Context131K
01 — Verdict

GPT-5.4 nano is our pick

GPT-5.4 nano is the better all-round choice, scoring 70/100 against DeepSeek V3.2 (60) and GPT OSS 120B (57). It leads on inputs & features and context window. The score weighs capability 67%, inputs & features 20%, context window 13%.

  • CapabilityGPT-5.4 nanoCapabilities Index (ECI): GPT-5.4 nano 146.3 · DeepSeek V3.2 146.3 · GPT OSS 120B 140.2
  • Lowest priceGPT OSS 120BGPT OSS 120B $0.263 · DeepSeek V3.2 $0.315 · GPT-5.4 nano $0.463 per 1M tokens (3:1 blend)
  • Longest contextGPT-5.4 nanoGPT-5.4 nano 400,000 · GPT OSS 120B 131,072 · DeepSeek V3.2 128,000 tokens
  • Widest inputsGPT-5.4 nanoDeepSeek V3.2: Text · GPT-5.4 nano: Text, Images · GPT OSS 120B: Text
  • Self-hostingDeepSeek V3.2 and GPT OSS 120BPublishes downloadable weights (MIT License)
How the score is built
MeasureWeightDeepSeek V3.2GPT-5.4 nanoGPT OSS 120B
CapabilityCapabilities Index (ECI)67%747468
Inputs & features20%357035
Context window13%325032
Overall100%60/10070/10057/100
02 — Side by side

Every spec in one table

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

DeepSeek V3.2 vs GPT-5.4 nano vs GPT OSS 120B specifications side by side
SpecificationDeepSeek V3.2DeepSeekGPT-5.4 nanoOpenAIGPT OSS 120BOpenAI
Capability
Capabilities Index (ECI)146.3146.3 (best)140.2
ECI rank#87 of 158#86 of 158 (best)#115 of 158
GPQA DiamondGraduate-level science questions83.4% (best)78.5%75.8%
FrontierMath Tiers 1–3Research-level mathematics—44.9%—
OTIS Mock AIME 2024–2025Competition mathematics87.8%87.8%88.9% (best)
SimpleQA VerifiedShort factual questions—11.7%—
Price per million tokens
Input$0.28$0.20$0.15 (best)
Output$0.42 (best)$1.25$0.60
Cached input—$0.02—
Blended (3:1)$0.315$0.463$0.263 (best)
Long-context rateSame rateSame rateSame rate
Price sourceMedian of 30 providersOfficial OpenAI APIMedian of 77 providers
Limits
Context window128,000 tokens400,000 tokens (best)131,072 tokens
Max output64,000 tokens128,000 tokens131,072 tokens (best)
Inputs and features
TextYesYesYes
ImagesNoYesNo
PDFsNoNoNo
AudioNoNoNo
VideoNoNoNo
ReasoningYesYeslow · medium · high · xhighYes
Tool callingYesYesYes
Structured outputNoYesNo
Availability
WeightsOpenMIT LicenseProprietaryOpen
API model ID—gpt-5.4-nano—
API providers303680 (best)
ReleasedDec 1, 2025Mar 17, 2026Aug 5, 2025
Knowledge cutoffJul 2024Aug 31, 2025Jun 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.

  • DeepSeek V3.2$3.64
  • GPT-5.4 nano$4.50
  • GPT OSS 120B$2.70
04 — Questions

Which should you choose?

Which is better: DeepSeek V3.2, GPT-5.4 nano or GPT OSS 120B?

GPT-5.4 nano is the better all-round choice, scoring 70/100 against DeepSeek V3.2 (60) and GPT OSS 120B (57). It leads on inputs & features and context window. The score weighs capability 67%, inputs & features 20%, context window 13%.

Which is cheaper, DeepSeek V3.2, GPT-5.4 nano or GPT OSS 120B?

GPT OSS 120B is cheaper at $0.15 input / $0.60 output per million tokens (median across 77 API providers). DeepSeek V3.2 costs $0.28 input / $0.42 output per million tokens (median across 30 API providers); GPT-5.4 nano costs $0.20 input / $1.25 output per million tokens (official OpenAI API price). At a typical mix of three input tokens to one output token, that is $0.263 per million tokens for GPT OSS 120B versus $0.315 for DeepSeek V3.2 (1.2× as much) and $0.463 for GPT-5.4 nano (1.8× as much).

Which scores higher on benchmarks?

GPT-5.4 nano scores higher on the Capabilities Index (ECI): GPT-5.4 nano 146.3 (#86 of 158), DeepSeek V3.2 146.3 (#87 of 158) and GPT OSS 120B 140.2 (#115 of 158). The confidence ranges of the top two overlap (144.7–148.0 vs 144.8–147.8), so treat the gap as small. On individual benchmarks: GPQA Diamond — DeepSeek V3.2 83.4%, GPT-5.4 nano 78.5%, GPT OSS 120B 75.8%; OTIS Mock AIME 2024–2025 — GPT OSS 120B 88.9%, DeepSeek V3.2 87.8%, GPT-5.4 nano 87.8%.

Which is better for coding?

There are no published SWE-bench Verified results for DeepSeek V3.2, GPT-5.4 nano and GPT OSS 120B yet, so there is no like-for-like coding score. On overall capability, GPT-5.4 nano 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 131,072 for GPT OSS 120B and 128,000 for DeepSeek V3.2. Maximum output per response: DeepSeek V3.2 up to 64,000, GPT-5.4 nano up to 128,000, GPT OSS 120B up to 131,072 tokens.

Which can read images, PDFs, audio or video?

DeepSeek V3.2 accepts text; GPT-5.4 nano accepts text and images; GPT OSS 120B accepts text. GPT-5.4 nano handles the widest range of inputs.

Are any of these open source?

DeepSeek V3.2 and GPT OSS 120B publishes its weights (MIT License) and can be self-hosted; GPT-5.4 nano is proprietary.

Which is newer?

GPT-5.4 nano is the newest, released Mar 17, 2026. DeepSeek V3.2 came out Dec 1, 2025; GPT OSS 120B came out Aug 5, 2025. Knowledge cutoff: DeepSeek V3.2 Jul 2024, GPT-5.4 nano Aug 31, 2025, GPT OSS 120B Jun 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.