ZMIME
Comparison · 3 models · Updated Nov 18, 2026

DeepSeek-V3.1 vs GPT OSS 120B vs Llama 4 Maverick 17B Instruct

Too close to call on our weighted score (Llama 4 Maverick 17B Instruct 59, DeepSeek-V3.1 57, GPT OSS 120B 57). The right pick depends on what you value most.

Price in rating: Excluded · Include price

  1. DeepSeek

    DeepSeek-V3.1

    Released Aug 21, 2025

    57/100
    • ECI140.8
    • Price$0.27 / $1.00
    • Context131K
  2. OpenAI

    GPT OSS 120B

    Released Aug 5, 2025

    57/100
    • ECI140.2
    • Price$0.15 / $0.60
    • Context131K
  3. Meta

    Llama 4 Maverick 17B Instruct

    Released Apr 5, 2025

    59/100
    • ECI132.2
    • Price$0.27 / $0.86
    • Context1M
01 — Verdict

Too close to call

It is close. Our weighted score puts them within 2 points (Llama 4 Maverick 17B Instruct 59/100, DeepSeek-V3.1 57/100, GPT OSS 120B 57/100), so choose by what matters most for your work: DeepSeek-V3.1 for raw capability, GPT OSS 120B on price and Llama 4 Maverick 17B Instruct for long inputs. The score weighs capability 67%, inputs & features 20%, context window 13%.

  • CapabilityDeepSeek-V3.1Capabilities Index (ECI): DeepSeek-V3.1 140.8 · GPT OSS 120B 140.2 · Llama 4 Maverick 17B Instruct 132.2
  • Lowest priceGPT OSS 120BGPT OSS 120B $0.263 · Llama 4 Maverick 17B Instruct $0.417 · DeepSeek-V3.1 $0.453 per 1M tokens (3:1 blend)
  • Longest contextLlama 4 Maverick 17B InstructLlama 4 Maverick 17B Instruct 1,000,000 · DeepSeek-V3.1 131,072 · GPT OSS 120B 131,072 tokens
  • Widest inputsLlama 4 Maverick 17B InstructDeepSeek-V3.1: Text · GPT OSS 120B: Text · Llama 4 Maverick 17B Instruct: Text, Images
  • Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
How the score is built
MeasureWeightDeepSeek-V3.1GPT OSS 120BLlama 4 Maverick 17B Instruct
CapabilityCapabilities Index (ECI)67%696861
Inputs & features20%353550
Context window13%323264
Overall100%57/10057/10059/100
02 — Side by side

Every spec in one table

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

DeepSeek-V3.1 vs GPT OSS 120B vs Llama 4 Maverick 17B Instruct specifications side by side
SpecificationDeepSeek-V3.1DeepSeekGPT OSS 120BOpenAILlama 4 Maverick 17B InstructMeta
Capability
Capabilities Index (ECI)140.8 (best)140.2132.2
ECI rank#112 of 158 (best)#115 of 158#134 of 158
GPQA DiamondGraduate-level science questions—75.8% (best)67.0%
OTIS Mock AIME 2024–2025Competition mathematics—88.9% (best)20.6%
Price per million tokens
Input$0.27$0.15 (best)$0.27
Output$1.00$0.60 (best)$0.86
Cached input———
Blended (3:1)$0.453$0.263 (best)$0.417
Long-context rateSame rateSame rateSame rate
Price sourceMedian of 18 providersMedian of 77 providersMedian of 16 providers
Limits
Context window131,072 tokens131,072 tokens1,000,000 tokens (best)
Max output8,192 tokens131,072 tokens (best)16,384 tokens
Inputs and features
TextYesYesYes
ImagesNoNoYes
PDFsNoNoNo
AudioNoNoNo
VideoNoNoNo
ReasoningYesYesNo
Tool callingYesYesYes
Structured outputNoNoNo
Availability
WeightsOpenMIT LicenseOpenOpen
API model ID———
API providers1880 (best)16
ReleasedAug 21, 2025Aug 5, 2025Apr 5, 2025
Knowledge cutoff—Jun 2024Aug 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-V3.1$4.70
  • GPT OSS 120B$2.70
  • Llama 4 Maverick 17B Instruct$4.42
04 — Questions

Which should you choose?

Which is better: DeepSeek-V3.1, GPT OSS 120B or Llama 4 Maverick 17B Instruct?

It is close. Our weighted score puts them within 2 points (Llama 4 Maverick 17B Instruct 59/100, DeepSeek-V3.1 57/100, GPT OSS 120B 57/100), so choose by what matters most for your work: DeepSeek-V3.1 for raw capability, GPT OSS 120B on price and Llama 4 Maverick 17B Instruct for long inputs. The score weighs capability 67%, inputs & features 20%, context window 13%.

Which is cheaper, DeepSeek-V3.1, GPT OSS 120B or Llama 4 Maverick 17B Instruct?

GPT OSS 120B is cheaper at $0.15 input / $0.60 output per million tokens (median across 77 API providers). Llama 4 Maverick 17B Instruct costs $0.27 input / $0.86 output per million tokens (median across 16 API providers); DeepSeek-V3.1 costs $0.27 input / $1.00 output per million tokens (median across 18 API providers). At a typical mix of three input tokens to one output token, that is $0.263 per million tokens for GPT OSS 120B versus $0.417 for Llama 4 Maverick 17B Instruct (1.6× as much) and $0.453 for DeepSeek-V3.1 (1.7× as much).

Which scores higher on benchmarks?

DeepSeek-V3.1 scores higher on the Capabilities Index (ECI): DeepSeek-V3.1 140.8 (#112 of 158), GPT OSS 120B 140.2 (#115 of 158) and Llama 4 Maverick 17B Instruct 132.2 (#134 of 158). The confidence ranges of the top two overlap (137.6–143.9 vs 138.3–142.1), so treat the gap as small.

Which is better for coding?

There are no published SWE-bench Verified results for DeepSeek-V3.1, GPT OSS 120B and Llama 4 Maverick 17B Instruct yet, so there is no like-for-like coding score. On overall capability, DeepSeek-V3.1 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?

Llama 4 Maverick 17B Instruct has the largest context window at 1,000,000 tokens, against 131,072 for DeepSeek-V3.1 and 131,072 for GPT OSS 120B. Maximum output per response: DeepSeek-V3.1 up to 8,192, GPT OSS 120B up to 131,072, Llama 4 Maverick 17B Instruct up to 16,384 tokens.

Which can read images, PDFs, audio or video?

DeepSeek-V3.1 accepts text; GPT OSS 120B accepts text; Llama 4 Maverick 17B Instruct accepts text and images. Llama 4 Maverick 17B Instruct handles the widest range of inputs.

Are any of these open source?

Yes, all three publish their weights (MIT License), so you can self-host them.

Which is newer?

DeepSeek-V3.1 is the newest, released Aug 21, 2025. GPT OSS 120B came out Aug 5, 2025; Llama 4 Maverick 17B Instruct came out Apr 5, 2025. Knowledge cutoff: GPT OSS 120B Jun 2024, Llama 4 Maverick 17B Instruct Aug 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.