ByteDance Doubao Seed 2.0 Code Preview — code-specialist variant with tiered pricing.
Doubao Seed 2.0 Code Preview (260215) is TheRouter's ByteDance/BytePlus route for the February 2026 code-specialized Seed 2.0 snapshot. It is exposed as bytedance/doubao-seed-2.0-code-preview-260215, a text-in/text-out OpenAI-compatible chat model with a 262,144-token context window, 8,192-token maximum completion budget, tool calling, JSON response formatting, and a higher price band above 128K prompt tokens.
Treat this page as the exact BytePlus / ModelArk-facing snapshot, not a floating Doubao coding alias. ByteDance's Seed2.0 announcement says the family includes a dedicated Code model for developers, while TheRouter's catalog pins the preview snapshot id and its token economics. For operators, it belongs in coding-agent, repository-analysis, refactor-review, and code-generation routes where exact snapshot control and 256K prompt capacity matter more than multimodal breadth.
BytePlus / ModelArk exposes the upstream doubao-seed-2-0-code-preview-260215 route as part of its hosted model surface, with prompt-length pricing tiers recorded in the official rate-card documentation.
TheRouter exposes the same snapshot as an OpenAI-compatible chat-completions model and records the base plus long-context token prices in standard-models.yaml. Cache-hit billing is not exposed on this public model row.
| Type | Rate |
|---|---|
| Prompts up to 128K tokens | |
| Input | $0.540 / 1M tokens |
| Output | $3.24 / 1M tokens |
| Cached input | $0.540 / 1M tokens |
| Prompts over 128K tokens | |
| Input | $1.08 / 1M tokens |
| Output | $6.48 / 1M tokens |
| Cached input | $1.08 / 1M tokens |
Tiered pricing: when a request's prompt (input, cached tokens included) is over 128K tokens, every token in that request is billed at the long-context rates; exactly 128K is still the base rate. This model publishes no separate cached-input price: cached tokens are billed at the input rate of their group (the "Cached input" row shows that rate).
| TheRouter model id | bytedance/doubao-seed-2.0-code-preview-260215standard-models.yaml ↗ | verified |
| Upstream snapshot | doubao-seed-2-0-code-preview-260215provider-models/astraflow-bytedance.yaml ↗ | verified |
| Context window | 262,144 tokensstandard-models.yaml ↗ | verified |
| Maximum completion | 8,192 tokensstandard-models.yaml ↗ | verified |
| Input / output modalities | Text in; text outstandard-models.yaml ↗ | verified |
| Supported OpenAI-style parameters | temperature, max_tokens, top_p, tools, tool_choice, response_format, stopstandard-models.yaml ↗ | verified |
| TheRouter base price | $0.50 / 1M input tokens; $3.00 / 1M output tokensstandard-models.yaml ↗ | verified |
| Long-context tier | Prompts over 128,000 and up to 256K tokens: $1.00 / 1M input; $6.00 / 1M outputBytePlus ModelArk pricing ↗ | verified |
| Training cutoff | Not publicly disclosed | unknown |
| License / weights | Proprietary hosted API; weights not publicBytePlus ModelArk ↗ | verified |
Use the global api.therouter.ai endpoint shown below for new integrations; the legacy China accelerated endpoint is retired.
curl https://api.therouter.ai/v1/chat/completions -H "Content-Type: application/json" -H "Authorization: Bearer $THE_ROUTER_API_KEY" -d '{
"model": "bytedance/doubao-seed-2.0-code-preview-260215",
"messages": [
{"role": "user", "content": "Summarize the key points from this input."}
]
}'Minimal OpenAI-compatible request for code explanation, repository notes, or patch planning.
curl https://api.therouter.ai/v1/chat/completions \
-H "Authorization: Bearer $THEROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "bytedance/doubao-seed-2.0-code-preview-260215",
"messages": [
{"role": "user", "content": "Explain why this TypeScript generic is unsafe and suggest a patch."}
]
}'The local ByteDance catalog records the public Code Preview snapshot with $0.50/$3.00 per million tokens in the base prompt band and $1.00/$6.00 above 128K prompt tokens.
ByteDance's Seed2.0 announcement presents Pro, Lite, Mini, and a dedicated Code model as the production-oriented lineup for agent and developer workloads.
doubao/doubao-seed-2-0-code?No. They are related Seed 2.0 coding routes, but this page is the exact ByteDance / BytePlus 260215 Code Preview snapshot in the bytedance namespace. Use the exact id shown here in production configuration and evaluations.
The TheRouter catalog records a higher prompt-length band. For prompts over 128,000 tokens and up to the 256K range, input and output prices double to $1.00 and $6.00 per million tokens respectively.
This scheduled curation loop has no paid API budget. The snippet verifier intentionally requires an operator-set SNIPPET_VERIFY_APPROVED=1, so snippets remain unproven until a human approves and funds a live run.
| source | URL | retrieved | |
|---|---|---|---|
| TheRouter model id | standard-models.yaml ↗ | 2026-07-29 | verified |
| Upstream snapshot | provider-models/astraflow-bytedance.yaml ↗ | 2026-07-29 | verified |
| Context window | standard-models.yaml ↗ | 2026-07-29 | verified |
| Maximum completion | standard-models.yaml ↗ | 2026-07-29 | verified |
| Input / output modalities | standard-models.yaml ↗ | 2026-07-29 | verified |
| Supported OpenAI-style parameters | standard-models.yaml ↗ | 2026-07-29 | verified |
| TheRouter base price | standard-models.yaml ↗ | 2026-07-29 | verified |
| Long-context tier | BytePlus ModelArk pricing ↗ | 2026-07-29 | verified |
| Training cutoff | — | — | unknown |
| License / weights | BytePlus ModelArk ↗ | 2026-08-06 | verified |
| TheRouter records BytePlus Seed 2.0 Code Preview (260215) tiered pricing | BytePlus ModelArk pricing ↗ | 2026-07-29 | verified |
| ByteDance ships Seed2.0 with a dedicated Code model | seed.bytedance.com ↗ | 2026-08-06 | verified |
| Is this the same route as `doubao/doubao-seed-2-0-code`? | standard-models.yaml ↗ | 2026-07-29 | to verify |
| What happens above 128K prompt tokens? | BytePlus ModelArk pricing ↗ | 2026-07-29 | to verify |