Seed 1.8
Seed 1.8 is capable in multimodal tasks and behind the leaders in coding. Too few results yet to rate reasoning, agentic tasks, 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 Reasoning, Agentic, Safety, Long Context, Math, Multilingual, Instruction Following or Factuality.
Price
$0.25input$2.00outputper million tokens
From deepinfra · All prices
Evidence
41results on41benchmarks
- 40 vendor-reported
- 1 cross-referenced
From 2 sources · latest Nov 19, 2026 · How verification works
Seed 1.8 benchmark results
41 results on 41 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.
20.4% behind the leader4 of 6 ranked benchmarks measured
- 83.40Nov 19, 2026
- 87.70Nov 19, 2026
- 73.20Nov 19, 2026
- CharXiv (reasoning)71.40Nov 19, 2026CharXiv-R
37.4% behind the leader2 of 10 ranked benchmarks measured
- 79.50Nov 19, 2026
- 72.90Nov 19, 2026
0 of 7 ranked benchmarks measured
- OSWorld-Verified61.90Nov 19, 2026OSWorld
- 67.60Nov 19, 2026
0 of 3 ranked benchmarks measured
- 66.70Nov 19, 2026
More results
Benchmarks outside the capability baskets. They are not ranked against a leader.
- 11.00Nov 19, 2026
- 63.80Nov 19, 2026
- 67.80Nov 19, 2026
- 82.70Nov 19, 2026
- 71.50Nov 19, 2026
- 60.80Nov 19, 2026
Show 24 more resultsHide 24 results
- 89.10Nov 19, 2026
- 94.30Nov 19, 2026
- 70.70Nov 19, 2026
- 77.00Nov 19, 2026
- 57.20Nov 19, 2026
- 81.30Nov 19, 2026
- 69.40Nov 19, 2026
- 61.50Nov 19, 2026
- 58.80Nov 19, 2026
- 63.90Nov 19, 2026
- 77.40Nov 19, 2026
- 73.00Nov 19, 2026
- 81.30Nov 19, 2026
- 62.40Nov 19, 2026
- 84.90Nov 19, 2026
- 73.10Nov 19, 2026
- 70.60Nov 19, 2026
- 42.00Nov 19, 2026
- OSWorld-Verified (Overall)61.90Oct 30, 2026
- 65.10Nov 19, 2026
- 56.30Nov 19, 2026
- 65.40Nov 19, 2026
- 64.80Nov 19, 2026
- 45.20Nov 19, 2026
Seed 1.8: common questions
Who makes Seed 1.8?
Seed 1.8 is made by ByteDance.
What is Seed 1.8 good at?
Seed 1.8 is capable in multimodal tasks and behind the leaders in coding. Too few results yet to rate reasoning, agentic tasks, safety, long context, math, multilingual tasks, instruction following, or factuality.
How much does Seed 1.8 cost?
Seed 1.8 costs $0.25 per million input tokens and $2.00 per million output tokens, according to deepinfra. At a mix of three input tokens to one output token, it is cheaper than 51% of the 336 priced models we track.
How many benchmarks has Seed 1.8 been tested on?
We track 41 results for Seed 1.8 on 41 benchmarks from 2 sources. The latest was recorded on Nov 19, 2026.
About this record
Where Seed 1.8's numbers come from, and every name it appears under.
- Tracked since
- Jun 22, 2026
- Newest source mention
- Sep 3, 2026
Where the results come from
Verification: 41 scores · 0 independently verified · 1 vendor cross-reference · 40 vendor-reported. How these tiers are assigned
From 2 sources on 2 sites. api.llm-stats.com supplies 40 of them. Bars are coloured by trust tier.
- api.llm-stats.com40
- huggingface.co1