Seed2.1 Turbo

Seed2.1 Turbo lists base rates of $0.50 per million input tokens and $2.50 per million output tokens. Estimate both input and output token volume and any additional charges for your workload. BytePlus ModelArk lists dola-seed-2-1-turbo at $0.50/$2.50 per MTok for prompts through 256K, with $0.10 cache-hit input and $0.0083 per MTok/hour standard cache storage. Batch and Flex are $0.25/$1.25 with $0.10 cache-hit input. Flex supports implicit caching only and has no additional cache-storage fee. These are BytePlus USD rates, not a verified Seed2.1 Pro or mainland Doubao tariff.

Seed2.1 Turbo accepts about 256K tokens of context. That only matters if you routinely send very long documents, large codebases, or multi-turn histories that approach that limit.

Seed2.1 Turbo is a balanced tier from ByteDance. It suits agents, coding, multimodal work. June 23 is the Seed2.1 family launch; the pinned SKU suffix is separate. BytePlus advertises 256K context, input and output including CoT, with 4K default output; these share one context budget and an exact integer output cap is unverified. thinking.type supports enabled (default) and disabled. All seven effort inputs are accepted, but this SKU's default/mapping cells are blank; do not promise seven distinct native levels. Summary reasoning tokens are billed using the original thinking tokens. Parameters, licence and weight availability remain unknown. Vendor model-card scores are separate from independent boards.