Skip to content
Metatext Decide — typed decisions for your agents, with a probability on every answerGet early access
Metatext
Sign Up

Aya Vision 32B

CohereOpen weights

Aya Vision 32B by Cohere costs — per 1M input tokens and — per 1M output tokens, with a 16K-token context window.

Pricing

Input (per 1M tokens)
—
Output (per 1M tokens)
—
Cached input (per 1M)
—

Specifications

Provider
Cohere
Context window
16K tokens
Parameters
—
Released
Mar 2025
Open weights
Yes
Frontier model
No

Benchmarks

LMArena Vision Elo
995.6

Speed & latency

Time to first token
2.32s
Output speed
36 tok/s

Fastest host: cohere (via hf-router).

Compare Aya Vision 32B with…

Aya Vision 32B vs DeepSeek V3.1 Terminus$1.00/1MAya Vision 32B vs Qwen2.5 Coder 32B Instruct$1.00/1MAya Vision 32B vs Nemotron 3 Super 120B A12B$1.00/1MAya Vision 32B vs MiniMax M2$1.02/1MAya Vision 32B vs MiniMax M3 Thinking$0.96/1MAya Vision 32B vs MiniMax M2.5$0.95/1M

FAQ

How much does Aya Vision 32B cost?

Aya Vision 32B is priced at — per 1M input tokens and — per 1M output tokens. Output tokens cost more than input.

What is Aya Vision 32B's context window?

Aya Vision 32B supports a context window of 16K tokens — the maximum prompt plus response length it can handle in a single request.

Is Aya Vision 32B open weight?

Yes — Aya Vision 32B has openly available weights.

Data as of 2026-10-21. Source: models.dev (open). Pricing is per 1M tokens (USD); confirm with the provider before production use.