Seed 2.1 Turbo
Seed 2.1 Turbo is at the frontier in multimodal tasks; and behind the leaders in coding, reasoning, and agentic tasks. Too few results yet to rate safety, long context, math, multilingual tasks, instruction following, or factuality.
Capability profile
Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.
Too few results yet to rate Safety, Long Context, Math, Multilingual, Instruction Following or Factuality.
Price
$0.50input$2.50outputper million tokens
From bytedance-seed · All prices
Evidence
52results on52benchmarks
- 1 independently verified
- 51 vendor-reported
From 2 sources · latest Nov 19, 2026 · How verification works
API features
Tool callingStructured outputsReasoning
As listed by OpenRouter
Seed 2.1 Turbo benchmark results
52 results on 52 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.
4.7% behind the leader4 of 6 ranked benchmarks measured
- 90.50Nov 19, 2026
- 89.00Nov 19, 2026
- 82.20Nov 19, 2026
- CharXiv (reasoning)83.60Nov 19, 2026CharXiv-R
Show 2 more multimodal resultsHide 2 multimodal results
- 79.40Nov 19, 2026
- 62.80Nov 19, 2026
28.7% behind the leader3 of 10 ranked benchmarks measured
- 57.80Nov 19, 2026
- 67.60Nov 19, 2026
- 57.00Nov 19, 2026
30.7% behind the leader2 of 6 ranked benchmarks measured
- 54.60Nov 19, 2026
- 61.30Nov 19, 2026
32.8% behind the leader3 of 7 ranked benchmarks measured
- 84.90Nov 19, 2026
- OSWorld-Verified76.40Nov 19, 2026OSWorld
- 80.30Nov 19, 2026
Show 1 more agentic resultHide 1 agentic result
- AA ApexAgents29.20Nov 19, 2026APEX-Agents
0 of 4 ranked benchmarks measured
- 29.70Nov 4, 2026
More results
Benchmarks outside the capability baskets. They are not ranked against a leader.
- MathArena Apex (Pass@1)35.40Nov 19, 2026MathArena Apex
- 80.60Nov 19, 2026
- 74.80Nov 19, 2026
- 77.20Nov 19, 2026
- 67.60Nov 19, 2026
- 83.10Nov 19, 2026
Show 30 more resultsHide 30 results
- 47.00Nov 19, 2026
- 62.90Nov 19, 2026
- 88.00Nov 19, 2026
- 70.90Nov 19, 2026
- 67.00Nov 19, 2026
- 23.00Nov 19, 2026
- 68.10Nov 19, 2026
- 82.50Nov 19, 2026
- 71.30Nov 19, 2026
- Finance Agent v1.156.00Nov 19, 2026
- 76.80Nov 19, 2026
- 92.70Nov 19, 2026
- 89.20Nov 19, 2026
- 65.90Nov 19, 2026
- 76.90Nov 19, 2026
- 31.40Nov 19, 2026
- 42.00Nov 19, 2026
- 43.70Nov 19, 2026
- 71.10Nov 19, 2026
- 69.70Nov 19, 2026
- 18.30Nov 19, 2026
- 49.40Nov 19, 2026
- 86.30Nov 19, 2026
- 71.10Nov 19, 2026
- 67.40Nov 19, 2026
- 56.80Nov 19, 2026
- 49.10Nov 19, 2026
- 71.40Nov 19, 2026
- 48.60Nov 19, 2026
- 11.00Nov 19, 2026
Seed 2.1 Turbo: common questions
Who makes Seed 2.1 Turbo?
Seed 2.1 Turbo is made by ByteDance.
What is Seed 2.1 Turbo good at?
Seed 2.1 Turbo is at the frontier in multimodal tasks; and behind the leaders in coding, reasoning, and agentic tasks. Too few results yet to rate safety, long context, math, multilingual tasks, instruction following, or factuality.
How much does Seed 2.1 Turbo cost?
Seed 2.1 Turbo costs $0.50 per million input tokens and $2.50 per million output tokens, according to bytedance-seed. At a mix of three input tokens to one output token, it costs more than 57% of the 336 priced models we track.
How many benchmarks has Seed 2.1 Turbo been tested on?
We track 52 results for Seed 2.1 Turbo on 52 benchmarks from 2 sources, 1 of them independently verified. The latest was recorded on Nov 19, 2026.
Which API features does Seed 2.1 Turbo support?
OpenRouter lists tool calling, structured outputs, and reasoning for Seed 2.1 Turbo.
About this record
Where Seed 2.1 Turbo's numbers come from, and every name it appears under.
- Tracked since
- Aug 13, 2026
- Newest source mention
- Nov 4, 2026
Where the results come from
Verification: 52 scores · 1 independently verified · 51 vendor-reported. How these tiers are assigned
From 2 sources on 2 sites. api.llm-stats.com supplies 51 of them; the 1 independently verified result comes from 1 site. Bars are coloured by trust tier.
- api.llm-stats.com51
- epoch.ai1