# Sources (accessed 2026-08-14):
# - https://seed.bytedance.com/en/seed2
# - https://www.volcengine.com/docs/82379/1330310
name = "Seed 1.6 Vision"
description = "ByteDance Seed multimodal model for image understanding, visual reasoning, and tool-assisted tasks"
family = "seed"
release_date = "2025-08-15"
last_updated = "2025-08-15"
attachment = true
reasoning = false
temperature = true
tool_call = true
structured_output = true
open_weights = false

[limit]
context = 256_000
output = 32_000

[modalities]
input = ["text", "image"]
output = ["text"]