ZMIME
Comparison · 3 models · Updated Nov 18, 2026

Aya Expanse 32B vs GPT-4o vs Hermes 3 70B Instruct

No independent capability benchmark covers every model here yet. The score uses price, inputs & features and context window.

Price in rating: Included · Exclude price

  1. Cohere

    Aya Expanse 32B

    Released Oct 24, 2024

    33/100
    • ECI—
    • Price$0.50 / $1.50
    • Context128K
  2. OpenAI

    GPT-4o

    Released May 13, 2024

    37/100
    • ECI128.1
    • Price$2.50 / $10.00
    • Context128K
  3. Nous Research

    Hermes 3 70B Instruct

    Released Aug 18, 2024

    34/100
    • ECI—
    • Price$0.70 / $0.70
    • Context131K
01 — Verdict

Capability evidence is incomplete

There is not enough shared capability evidence to name an overall winner. The scores (GPT-4o 37/100, Hermes 3 70B Instruct 34/100, Aya Expanse 32B 33/100) compare the available specifications and price. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.

  • CapabilityNot enough dataNo independent benchmark covers every model here yet
  • Lowest priceHermes 3 70B InstructHermes 3 70B Instruct $0.70 · Aya Expanse 32B $0.75 · GPT-4o $4.38 per 1M tokens (3:1 blend)
  • Longest contextHermes 3 70B InstructHermes 3 70B Instruct 131,072 · Aya Expanse 32B 128,000 · GPT-4o 128,000 tokens
  • Widest inputsGPT-4oAya Expanse 32B: Text · GPT-4o: Text, Images, PDFs · Hermes 3 70B Instruct: Text
  • Self-hostingAya Expanse 32B and Hermes 3 70B InstructPublishes downloadable weights (CC-BY-NC-4.0)
How the score is built
MeasureWeightAya Expanse 32BGPT-4oHermes 3 70B Instruct
Price50%531955
Inputs & features30%0700
Context window20%323232
Specs & price only100%33/10037/10034/100

Left out because at least one model lacks the data: capability. The remaining weights were rescaled.

02 — Side by side

Every spec in one table

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

Aya Expanse 32B vs GPT-4o vs Hermes 3 70B Instruct specifications side by side
SpecificationAya Expanse 32BCohereGPT-4oOpenAIHermes 3 70B InstructNous Research
Capability
Capabilities Index (ECI)—128.1—
ECI rank—#139 of 158—
GPQA DiamondGraduate-level science questions—48.9%—
OTIS Mock AIME 2024–2025Competition mathematics—6.3%—
Price per million tokens
Input$0.50 (best)$2.50$0.70
Output$1.50$10.00$0.70 (best)
Cached input—$1.25—
Blended (3:1)$0.75$4.38$0.70 (best)
Long-context rateSame rateSame rateSame rate
Price sourceOfficial Cohere APIOfficial OpenAI APIMedian of 1 providers
Limits
Context window128,000 tokens128,000 tokens131,072 tokens (best)
Max output4,000 tokens16,384 tokens (best)16,384 tokens (best)
Inputs and features
TextYesYesYes
ImagesNoYesNo
PDFsNoYesNo
AudioNoNoNo
VideoNoNoNo
ReasoningNoNoNo
Tool callingNoYesNo
Structured outputNoYesNo
Availability
WeightsOpenCC-BY-NC-4.0ProprietaryOpen
API model IDc4ai-aya-expanse-32bgpt-4o—
API providers227 (best)1
ReleasedOct 24, 2024May 13, 2024Aug 18, 2024
Knowledge cutoff—Sep 2023Dec 31, 2023
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.

  • Aya Expanse 32B$8.00
  • GPT-4o$45.00
  • Hermes 3 70B Instruct$8.40
04 — Questions

Which should you choose?

Which is better: Aya Expanse 32B, GPT-4o or Hermes 3 70B Instruct?

There is not enough shared capability evidence to name an overall winner. The scores (GPT-4o 37/100, Hermes 3 70B Instruct 34/100, Aya Expanse 32B 33/100) compare the available specifications and price. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.

Which is cheaper, Aya Expanse 32B, GPT-4o or Hermes 3 70B Instruct?

Hermes 3 70B Instruct is cheaper at $0.70 input / $0.70 output per million tokens (median across 1 API provider). Aya Expanse 32B costs $0.50 input / $1.50 output per million tokens (official Cohere API price); GPT-4o costs $2.50 input / $10.00 output per million tokens (official OpenAI API price). At a typical mix of three input tokens to one output token, that is $0.70 per million tokens for Hermes 3 70B Instruct versus $0.75 for Aya Expanse 32B (1.1× as much) and $4.38 for GPT-4o (6.3× as much).

Which scores higher on benchmarks?

There is no independent benchmark that covers all three models yet. Aya Expanse 32B has not been scored yet, GPT-4o has an ECI of 128.1 and Hermes 3 70B Instruct has not been scored yet.

Which is better for coding?

There are no published SWE-bench Verified results for Aya Expanse 32B, GPT-4o and Hermes 3 70B Instruct yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. Note that Aya Expanse 32B and Hermes 3 70B Instruct does not support tool calling, which most coding agents need.

Which has the bigger context window?

Hermes 3 70B Instruct has the largest context window at 131,072 tokens, against 128,000 for Aya Expanse 32B and 128,000 for GPT-4o. Maximum output per response: Aya Expanse 32B up to 4,000, GPT-4o up to 16,384, Hermes 3 70B Instruct up to 16,384 tokens.

Which can read images, PDFs, audio or video?

Aya Expanse 32B accepts text; GPT-4o accepts text, images and PDFs; Hermes 3 70B Instruct accepts text. GPT-4o handles the widest range of inputs.

Are any of these open source?

Aya Expanse 32B and Hermes 3 70B Instruct publishes its weights (CC-BY-NC-4.0) and can be self-hosted; GPT-4o is proprietary.

Which is newer?

Aya Expanse 32B is the newest, released Oct 24, 2024. Hermes 3 70B Instruct came out Aug 18, 2024; GPT-4o came out May 13, 2024. Knowledge cutoff: GPT-4o Sep 2023, Hermes 3 70B Instruct Dec 31, 2023.

How do you decide the winner?

This rating uses price (50%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (30%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (20%, 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.