whatmodel.sh / Doubao Seed 2.1 Turbo (260628)
Doubao Seed 2.1 Turbo (260628)
↗ share
+ contribute
Doubao Seed 2.1 Turbo (260628) is a closed-weights language model from ByteDance Seed, released June 2026. Licence: proprietary.
Catalogue identity notes
Exact Volcengine Ark model id 'doubao-seed-2-1-turbo-260628' (model list, observed 2026-11-12). Released month from the id suffix and the Seed2.1 family release (blog 2026-06-23); the blog itself does not name Turbo. No existing database identity. Matches Epoch version 'doubao-seed-2.1-turbo-260628', release date 2026-06-23.
Model information sources
Score within the LMArena (text) comparison set only.
$ how it compares
The LLMs ranked either side of Doubao Seed 2.1 Turbo (260628), out of 897 with a score, then more from ByteDance Seed.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 84 | 70±1 | local | |||
| 85 | 70±2 | api $1.78/M | |||
| 86 | 70±2 | api $0.89/M | |||
| 87 | 70±1 | api $10.00/M | |||
| 88 | 70±5 | api price unknown | |||
| more from ByteDance Seed | |||||
| 90 | 70±1 | api $0.95/M | |||
| 100 | 69±1 | api $0.20/M | |||
| 220 | 62±1 | local | |||
$ what it's good at
math▁▁▃▅█▆▅█▄▁▁▁ #57 of 440 * · 9.5% of its score · 1 test
Beats GLM-5.3-Flash, LongCat 2.0 (1.6T-A48B), Claude Sonnet 4.6 here, though they rank higher overall. Trails A.X K2 (688B-A33B) here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| OTIS mock AIME 2024–2025 | 96.1% ✎ | ▃▃▄▃▃▃▇▆█▄▃▆ beats 170/207 | 9.5% |
other tests3 not in a task · 89% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Chess Puzzles (Epoch) | 26.0% ✎ | █▁▁▁▁▂▄▃▃▃▂▁ beats 99/157 | 41% |
| GPQA Diamond | 89.6% ✎ | ▁▁▁▁▂▃▅▄▄▆█▃ beats 174/221 | 33% |
| SimpleQA Verified | 29.7% ✎ | ▁▂▂▂▅█▄▃▂··▁ beats 20/91 | 15% |
$ where to run it
Through an API
api · Volcengine Ark · $0.89/M · $0.4453 in / $2.22651 out per 1M tokens · 256k context · docs.volcengine.com ↗ · price sourceUSD per 1M tokens from models.dev (CNY converted at 6.737012, 2026-08-26); cache read excluded. Volcengine pricing page (observed 2026-11-12): doubao-seed-2.1-turbo 'input [0, 256]K: 3.00 / cache hit 0.60 / output 15.00' CNY per 1M tokens, consistent.
$ how the score was worked out · Epoch AI — internal evaluatio… 99%
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
| 4 | 99% | |
| prior · pull toward the middle | 1% | |
| weighted average | = 70.3 |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json