ZMIME
Comparison · 3 models · Updated Nov 2, 2026

Gemma 3 27B IT vs GPT OSS 20B vs Llama 4 Scout 17B Instruct

Too close to call on our weighted score (Llama 4 Scout 17B Instruct 61, GPT OSS 20B 61, Gemma 3 27B IT 57). The right pick depends on what you value most.

  1. Google

    Gemma 3 27B IT

    Released Mar 12, 2025

    57/100
    • ECI129.9
    • Price$0.11 / $0.30
    • Context131K
    Full specs Remove Gemma 3 27B IT

  2. OpenAI

    GPT OSS 20B

    Released Aug 5, 2025

    61/100
    • ECI137.8
    • Price$0.07 / $0.25
    • Context131K
    Full specs Remove GPT OSS 20B

  3. Meta

    Llama 4 Scout 17B Instruct

    Released Apr 5, 2025

    61/100
    • ECI129.6
    • Price$0.17 / $0.59
    • Context10M
    Full specs Remove Llama 4 Scout 17B Instruct
01 — Verdict

Too close to call

It is close. Our weighted score puts them within a point (Llama 4 Scout 17B Instruct 61/100, GPT OSS 20B 61/100, Gemma 3 27B IT 57/100), so choose by what matters most for your work: GPT OSS 20B for raw capability and Llama 4 Scout 17B Instruct for long inputs. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.

  • CapabilityGPT OSS 20BCapabilities Index (ECI): GPT OSS 20B 137.8 · Gemma 3 27B IT 129.9 · Llama 4 Scout 17B Instruct 129.6
  • Lowest priceGPT OSS 20BGPT OSS 20B $0.115 · Gemma 3 27B IT $0.157 · Llama 4 Scout 17B Instruct $0.275 per 1M tokens (3:1 blend)
  • Longest contextLlama 4 Scout 17B InstructLlama 4 Scout 17B Instruct 10,000,000 · Gemma 3 27B IT 131,072 · GPT OSS 20B 131,072 tokens
  • Widest inputsGemma 3 27B IT and Llama 4 Scout 17B InstructGemma 3 27B IT: Text, Images · GPT OSS 20B: Text · Llama 4 Scout 17B Instruct: Text, Images
  • Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
How the score is built
MeasureWeightGemma 3 27B ITGPT OSS 20BLlama 4 Scout 17B Instruct
CapabilityCapabilities Index (ECI)50%526151
Price25%849073
Inputs & features15%503550
Context window10%2525100
Overall100%57/10061/10061/100
02 — Side by side

Every spec in one table

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

Gemma 3 27B IT vs GPT OSS 20B vs Llama 4 Scout 17B Instruct specifications side by side
SpecificationGemma 3 27B ITGoogleGPT OSS 20BOpenAILlama 4 Scout 17B InstructMeta
Capability
Capabilities Index (ECI)129.9137.8 (best)129.6
ECI rank#137 of 165#122 of 165 (best)#138 of 165
GPQA DiamondGraduate-level science questions47.7%60.8% (best)51.8%
OTIS Mock AIME 2024–2025Competition mathematics22.5%65.3% (best)7.8%
Price per million tokens
Input$0.11$0.07 (best)$0.17
Output$0.30$0.25 (best)$0.59
Cached input———
Blended (3:1)$0.157$0.115 (best)$0.275
Long-context rateSame rateSame rateSame rate
Price sourceMedian of 16 providersMedian of 37 providersMedian of 13 providers
Limits
Context window131,072 tokens131,072 tokens10,000,000 tokens (best)
Max output131,072 tokens (best)32,768 tokens16,384 tokens
Inputs and features
TextYesYesYes
ImagesYesNoYes
PDFsNoNoNo
AudioNoNoNo
VideoNoNoNo
ReasoningNoYesNo
Tool callingYesYesYes
Structured outputNoNoNo
Availability
WeightsOpenOpenOpen
API model ID———
API providers1741 (best)13
ReleasedMar 12, 2025Aug 5, 2025Apr 5, 2025
Knowledge cutoffAug 2024Jun 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.

  • Gemma 3 27B IT$1.69
  • GPT OSS 20B$1.20
  • Llama 4 Scout 17B Instruct$2.88
04 — Questions

Which should you choose?

Which is better: Gemma 3 27B IT, GPT OSS 20B or Llama 4 Scout 17B Instruct?

It is close. Our weighted score puts them within a point (Llama 4 Scout 17B Instruct 61/100, GPT OSS 20B 61/100, Gemma 3 27B IT 57/100), so choose by what matters most for your work: GPT OSS 20B for raw capability and Llama 4 Scout 17B Instruct for long inputs. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.

Which is cheaper, Gemma 3 27B IT, GPT OSS 20B or Llama 4 Scout 17B Instruct?

GPT OSS 20B is cheaper at $0.07 input / $0.25 output per million tokens (median across 37 API providers). Gemma 3 27B IT costs $0.11 input / $0.30 output per million tokens (median across 16 API providers); Llama 4 Scout 17B Instruct costs $0.17 input / $0.59 output per million tokens (median across 13 API providers). At a typical mix of three input tokens to one output token, that is $0.115 per million tokens for GPT OSS 20B versus $0.157 for Gemma 3 27B IT (1.4× as much) and $0.275 for Llama 4 Scout 17B Instruct (2.4× as much).

Which scores higher on benchmarks?

GPT OSS 20B scores higher on the Capabilities Index (ECI): GPT OSS 20B 137.8 (#122 of 165), Gemma 3 27B IT 129.9 (#137 of 165) and Llama 4 Scout 17B Instruct 129.6 (#138 of 165). Their confidence ranges do not overlap (133.1–139.7 vs 124.2–132.1), so the gap is a real one. On individual benchmarks: GPQA Diamond — GPT OSS 20B 60.8%, Llama 4 Scout 17B Instruct 51.8%, Gemma 3 27B IT 47.7%; OTIS Mock AIME 2024–2025 — GPT OSS 20B 65.3%, Gemma 3 27B IT 22.5%, Llama 4 Scout 17B Instruct 7.8%.

Which is better for coding?

There are no published SWE-bench Verified results for Gemma 3 27B IT, GPT OSS 20B and Llama 4 Scout 17B Instruct yet, so there is no like-for-like coding score. On overall capability, GPT OSS 20B 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 Scout 17B Instruct has the largest context window at 10,000,000 tokens, against 131,072 for Gemma 3 27B IT and 131,072 for GPT OSS 20B. Maximum output per response: Gemma 3 27B IT up to 131,072, GPT OSS 20B up to 32,768, Llama 4 Scout 17B Instruct up to 16,384 tokens.

Which can read images, PDFs, audio or video?

Gemma 3 27B IT accepts text and images; GPT OSS 20B accepts text; Llama 4 Scout 17B Instruct accepts text and images. Gemma 3 27B IT handles the widest range of inputs.

Are any of these open source?

Yes, all three publish their weights, so you can self-host them.

Which is newer?

GPT OSS 20B is the newest, released Aug 5, 2025. Llama 4 Scout 17B Instruct came out Apr 5, 2025; Gemma 3 27B IT came out Mar 12, 2025. Knowledge cutoff: Gemma 3 27B IT Aug 2024, GPT OSS 20B Jun 2024, Llama 4 Scout 17B Instruct Aug 2024.

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 Nov 2, 2026.

05 — Keep comparing

Related comparisons