ByteDanceLanguage modelDeprecated
Seed 1.8
Seed 1.8 has no overall index value yet because there are too few independent benchmark results. From the vendor-reported results we estimate about 69.3 points (95% range 50.8 to 87.9). This is not an independent measurement.
n/aOverall indexhigher is better
n/ano measurement yet
Input $0.25 · Output $2
$0.69per 1M tokenslower is better
256Ktokens at once≈ 380 A4 pages
17out of 100same as last month
Summary
It is fairly cheap: $0.69 per 1M tokens (input and output blended); a typical model costs $1.56.
Seed 1.8 combines three reasoning modes with image understanding and a 256K-token context. ByteDance presents the proprietary service as a generalized agent model for search, coding, and GUI tasks. ByteDance publishes no parameter count, so the accessible model form is defined by its product role and tool surface.
Seed 2.0 Pro is documented as the direct development after it and is positioned by ByteDance as stronger on visual, logical, and unstructured tasks. Seed 1.8 remains a proprietary API service without publicly released weights.
All benchmark results
Every published result with its source, test setup and comparison models.
Show all 138 published benchmark results
Page 1 of 12
Measurements without matching peer values
These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.
BrowseComp-en (Show measurement, test conditions, and source)
- Source value
- 67.6
- Score
- 67.6
- Metric
- percent
- Unit
- %
- Category
- search
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- ByteDance Seed official benchmark table
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. official release prose; exact score stated in text
- Context
- The release page prose states the score; surrounding comparison charts were not used.
VLMsAreBiased (Show measurement, test conditions, and source)
- Source value
- 62
- Score
- 62
- Metric
- percent
- Unit
- %
- Category
- visual-knowledge
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- ByteDance Seed official benchmark table
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. official release prose; exact score stated in text
- Context
- The release page prose states the score; surrounding comparison charts were not used.
VideoMME (Show measurement, test conditions, and source)
- Source value
- 87.8
- Score
- 87.8
- Metric
- percent
- Unit
- %
- Category
- long-video
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- ByteDance Seed official benchmark table
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. official release prose; exact score stated in text
- Context
- The release page prose states the score; surrounding comparison charts were not used.
WorldTravel (Show measurement, test conditions, and source)
- Source value
- 47.2
- Score
- 47.2
- Metric
- percent
- Unit
- %
- Category
- multimodal-agent
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- ByteDance Seed official benchmark table
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. official release prose; exact score stated in text
- Context
- The release page prose states the score; surrounding comparison charts were not used.
ZeroBench (Show measurement, test conditions, and source)
- Source value
- 11
- Score
- 11
- Metric
- percent
- Unit
- %
- Category
- visual-reasoning
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- ByteDance Seed official benchmark table
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. official release prose; exact score stated in text
- Context
- The release page prose states the score; surrounding comparison charts were not used.
AIME-25 (Show measurement, test conditions, and source)
- Source value
- 94.3
- Score
- 94.3
- Metric
- Pass@1
- Unit
- %
- Category
- reasoning
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Seed1.8 Model Card, arXiv:2603.20633v3
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. Seed1.8, think-high, Pass@1, default no-tools setup, Table 1.
HMMT25 (Feb) (Show measurement, test conditions, and source)
- Source value
- 89.7
- Score
- 89.7
- Metric
- Pass@1
- Unit
- %
- Category
- reasoning
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Seed1.8 Model Card, arXiv:2603.20633v3
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. Seed1.8, think-high, Pass@1, default no-tools setup, Table 1.
BeyondAIME (Show measurement, test conditions, and source)
- Source value
- 77
- Score
- 77
- Metric
- Pass@1
- Unit
- %
- Category
- reasoning
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Seed1.8 Model Card, arXiv:2603.20633v3
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. Seed1.8, think-high, Pass@1, default no-tools setup, Table 1.
AMO-Bench (Show measurement, test conditions, and source)
- Source value
- 60
- Score
- 60
- Metric
- Pass@1
- Unit
- %
- Category
- reasoning
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Seed1.8 Model Card, arXiv:2603.20633v3
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. Seed1.8, think-high, Pass@1, default no-tools setup, Table 1.
IMO-AnswerBench (Show measurement, test conditions, and source)
- Source value
- 76.3
- Score
- 76.3
- Metric
- Pass@1
- Unit
- %
- Category
- reasoning
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Seed1.8 Model Card, arXiv:2603.20633v3
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. Seed1.8, think-high, Pass@1, code tools enabled as stated in the table, Table 1.
AetherCode (Show measurement, test conditions, and source)
- Source value
- 38.2
- Score
- 38.2
- Metric
- Pass@1
- Unit
- %
- Category
- coding
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Seed1.8 Model Card, arXiv:2603.20633v3
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. Seed1.8, think-high, Pass@1, default no-tools setup, Table 1.
LiveCodeBench v6 (Show measurement, test conditions, and source)
- Source value
- 79.5
- Score
- 79.5
- Metric
- Pass@1
- Unit
- %
- Category
- coding
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- Seed1.8 Model Card, arXiv:2603.20633v3
- Status
- active
- Retrieved at
- 2026-10-17
- Methodology
- Vendor-reported result. Seed1.8, think-high, Pass@1, default no-tools setup, Table 1.
Page 1 of 12
Popularity compared
How often the model is searched for, used and mentioned, compared with every model of its category.
Popularity index
The 100 most popular models of this category, 0 to 100, higher is more popular
20 of 101 models
As of November 2026 · Source: gradually.ai popularity index
Over time
Weekly values from October 18, 2026 to November 5, 2026
Seed 1.8Amazon Nova 2 LiteMinistral 8BGPT-5.4 ProQwen 2.5 72B
All specifications
Input and output
Watermarking and provenance
AI watermarking refers to invisible signals embedded in the output. Visible marks and provenance metadata are listed separately. Unknown means there is insufficient evidence for that particular field.
- AI watermarking
- Unknown
- no source
- Watermarking technique
- Not known
- no source
Licence and access
API
Further details
Prices
Shutdown announced for November 11, 2026. Rates last verified on September 22, 2026.
1 offer, prices in US dollars per 1M tokens, blended from 3 parts input and 1 part output.
| Provider | Input | Output | Cache read | Blended |
|---|---|---|---|---|
| ByteDancedirect | $0.25 | $2 | $0.05 | $0.69 |
| ByteDanceabove 128,000 context tokensdirect | $0.50 | $4 | n/a | $1.38 |
Cache prices come from the direct tariff; OpenRouter does not report them for the other providers. Direct price according to the developer, checked Sep 22, 2026. Open the LLM price index
Unchanged since tracking began on Aug 24, 2026.
InputOutput
Not included in any subscription in Germany.
Frequently asked questions about Seed 1.8
How good is Seed 1.8 in benchmarks?
Independent results are not yet enough for the overall index. An estimate from 7 vendor-reported results (GPQA Diamond 83.8%, LiveCodeBench v6 79.5%, MMLU-Pro 84.9%, MMMU-Pro 73.2%, MMMU (val) 83.4%, SWE-bench Verified 72.9%, Terminal-Bench 2.0 45.2%) gives about 69.3 points, with a 95% range of 50.8 to 87.9. The estimate is calibrated on models with both vendor-reported and independent results. Vendor numbers run clearly above independent measurements on some tests; the calibration offsets that on average.
Vendor estimate, not an overall index value
How much does Seed 1.8 cost?
Through the API $0.25 per 1M input tokens and $2 output. That is fairly cheap.
Prices from Oct 19, 2026
Aug 24, 2026Oct 19, 2026
Unchanged since tracking began on Aug 24, 2026.
How large is the context window of Seed 1.8?
256,000 tokens. That is how much text Seed 1.8 reads at once, about 380 A4 pages.
Page count as a rule of thumb from Artificial Analysis
Can I run Seed 1.8 locally?
No. ByteDance does not publish the weights.
Stated by ByteDance
- Open weights
- no
- License
- proprietary BytePlus terms
- Parameters
- n/a
- active per token
- n/a
Where can I use Seed 1.8?
Through the API. Subscription access is not documented in the available plan data.
Through the API
$0.25 input, $2 output per 1M tokens, unchanged since Aug 24, 2026.
Sources and data date
Show all 17 sources
Benchmark sources at a glance
Which independent evaluators tested this model. This is coverage, not a quality ranking.
No active result from the reviewed external evaluators can currently be attributed to this model. Other source evidence remains available below.
Without vendor claims and historical or disputed results; those stay labelled in the single values.
| Type | Evidence and data date |
|---|---|
| Price | Base API prices without caching or batch discounts. Higher context tiers and other rates are listed under Costs. |
| Input | Documented modalities. Retrieved September 8, 2026. |
| Output | Documented modalities. Retrieved September 8, 2026. |
| Research status | Research date September 8, 2026. 9 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication. |
| Not found in the inspected sources | parameters-billions Original wording No exact value located in the checked primary model card/API documentation. |
| Not found in the inspected sources | active-parameters-billions Original wording No exact value located in the checked primary model card/API documentation. |
| Not found in the inspected sources | knowledge-cutoff Original wording No exact value located in the checked primary model card/API documentation. |
| Unresolved | AA benchmark rows Original wording No exact Artificial Analysis model page located in the checked public sitemap. |
| Not published | immutable-provider-endpoint-revision Note The paper gives the exact Doubao-Seed product name and sampling settings but does not publish the underlying provider endpoint revision or immutable model digest. |
| Not published | FinReasoning endpoint and immutable serving revision Note The paper reports the product label Doubao-Seed-1.8 but does not disclose its exact API route or immutable serving revision. |
| Unresolved | unitAndDirection Note Table 1 does not state a unit or direction for the Seed-1.8 cells; these remain unknown. |
| Unresolved | sampleCount Note No per-model evaluated sample count is stated in the captured Table 1 source. |
| Unresolved | confidenceInterval Note No confidence interval is published for the Seed-1.8 Table 1 cells. |
| Additional source | Context window data · BytePlus Seed 1.8 documentation · Editorial description reviewed October 2, 2026 |
| Additional source | Seed1.8 Model Card, arXiv:2603.20633v3 (retrieved October 17, 2026) |
| Additional source | VisualNeedle: Benchmarking Active Visual Search in Information-Dense Scenes (arXiv:2605.26380v2) (retrieved October 17, 2026) |
| Additional source | FinReasoning benchmark paper (arXiv:2603.19254v2) (retrieved October 17, 2026) |
| Additional source | A benchmark for assessing large language models on molecular-to-food and food-to-molecular prediction tasks (Frontiers in Artificial Intelligence, DOI 10.3389/frai.2026.1818923) (retrieved October 17, 2026) |
| Additional source | Towards One-to-Many Temporal Grounding (arXiv:2606.06294) (retrieved October 19, 2026) |
| Additional source | The Aggregate, ByteDance company rating table (retrieved October 24, 2026) |
| Additional source | ByteDance Seed 1.8 release · Editorial description reviewed October 2, 2026 |
| Additional source | ByteDance Seed 2.0 launch · Editorial description reviewed October 2, 2026 |
| Additional source | OpenRouter: Popularity index (derived), CC BY 4.0 |
| Additional source | Vercel AI Gateway: Popularity index (derived), CC BY 4.0 |
| Additional source | Ollama: cumulative downloads for the popularity index |
| Additional source | Replicate: cumulative runs for the popularity index |
| Additional source | Google Trending Now: limited feed selection for Germany and the US |
| Additional source | Creative Commons Attribution 4.0 |