whatmodel.sh / Doubao Seed 2.0 Pro (260215)
Doubao Seed 2.0 Pro (260215)
↗ share
+ contribute
Doubao Seed 2.0 Pro (260215) is a closed-weights language model from ByteDance Seed, released February 2026. Licence: proprietary.
Catalogue identity notes
Exact Volcengine Ark model id 'doubao-seed-2-0-pro-260215'. ByteDance Seed's Seed2.0 page reports a separate column 'Seed2.0 Pro(0215)' (Seed2.0 model card path /seed2/0214/). Distinct from dola-seed-2-0-pro: that entry is Arena's 'dola-seed-2.0-pro' (Dola is the international app; the Seed2.0 page's BytePlus 'Try Now' link uses a different checkpoint, seed-2-0-pro-260328), and no maker page ties it to 260215. Matches Epoch version 'doubao-seed-2.0-pro-260215' (Epoch runs 2026-10-28/30, before EOS).
Model information sources
Score within the LMArena (text) comparison set only.
$ how it compares
The LLMs ranked either side of Doubao Seed 2.0 Pro (260215), out of 897 with a score, then more from ByteDance Seed.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 88 | 70±5 | api price unknown | |||
| 89 | 70±1 | local | |||
| 90 | 70±1 | api $0.95/M | |||
| 91 | 70±1 | api $1.71/M +local | |||
| 92 | 70±1 | api $1.56/M | |||
| more from ByteDance Seed | |||||
| 85 | 70±2 | api $1.78/M | |||
| 86 | 70±2 | api $0.89/M | |||
| 100 | 69±1 | api $0.20/M | |||
$ what it's good at
agent tasks█··········█ #1 of 2 · 24% of its score · 4 tests
| test | result | vs the others in its run | counts |
|---|---|---|---|
| SWE-bench Pro (split unspecified) lab | 46.9% ✎ | ▁·▁▁▁▁▁▄█▁▁▁ beats 28/64 | 9.0% |
| Terminal-Bench 2.0 lab | 55.8% ✎ | ▁▁▄▄▇█▇█▅▁▃▃ beats 24/38 | 5.8% |
| SWE-bench Multilingual lab | 71.7% ✎ | ▁··▂▂▃▆█▃▂▁▁ beats 42/69 | 5.2% |
| BrowseComp lab | 77.3% ✎ | ▁·▃▁▁▂▆▄█▃▂▄ beats 37/48 | 3.5% |
coding█··········█ #1 of 2 * · 14% of its score · 2 tests
| test | result | vs the others in its run | counts |
|---|---|---|---|
| SWE-bench Pro (split unspecified) lab | 46.9% ✎ | ▁·▁▁▁▁▁▄█▁▁▁ beats 28/64 | 9.0% |
| SWE-bench Multilingual lab | 71.7% ✎ | ▁··▂▂▃▆█▃▂▁▁ beats 42/69 | 5.2% |
math▁▁▃▅█▆▅█▄▁▁▁ #66 of 440 * · 4.1% of its score · 1 test
Beats Claude Sonnet 4.6, Qwen3.6 Max Preview, Qwen3.7 Plus here, though they rank higher overall. Trails A.X K2 (688B-A33B), Inkling Small (266B-A12B), DeepSeek V4 Flash (Preview, Apr 2026) here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| OTIS mock AIME 2024–2025 | 93.9% ✎ | ▃▃▄▃▃▃▇▆█▄▃▆ beats 158/207 | 4.1% |
vision & documents█··········█ #2 of 2 · 13% of its score · 3 tests
Trails Doubao Seed 2.0 Lite (260428) here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| CharXiv (reasoning questions) lab | 80.5% ✎ | ▂▄▃▄▅▅▅▅▅█▆▂ beats 80/105 | 4.6% |
| CharXiv descriptive questions lab | 93.5% ✎ | ▂▄▄▅▇▄██▇▃▂▂ beats 57/61 | 4.5% |
| MMMU Pro lab | 78.2% ✎ | ▁▂▁▂▅▃▃▆▂█▄▃ beats 47/55 | 3.7% |
web search█··········█ #1 of 2 * · 3.5% of its score · 1 test
| test | result | vs the others in its run | counts |
|---|---|---|---|
| BrowseComp lab | 77.3% ✎ | ▁·▃▁▁▂▆▄█▃▂▄ beats 37/48 | 3.5% |
other tests6 not in a task · 59% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| GPQA Diamond +1 | 88.2% ✎ | ▁▁▁▁▂▃▅▄▄▆█▃ beats 164/221 | 25% |
| Chess Puzzles (Epoch) | 20.0% ✎ | █▁▁▁▁▂▄▂▃▃▂▁ beats 78/157 | 16% |
| SimpleQA Verified | 39.4% ✎ | ▁▂▂▂▅█▄▃▂··▁ beats 36/91 | 5.9% |
| Humanity's Last Exam lab | 32.4% ✎ | ▁▁▁▄▆▇▅█▆▁▂▁ beats 85/115 | 5.6% |
| MultiChallenge lab | 68.3% ✎ | ▁·▁·▁▅█▃▅▃█▃ beats 32/33 | 3.1% |
| SuperGPQA lab | 68.7% ✎ | ▂▁▂▂▅▆▃▇▇▇█▆ beats 46/48 | 3.0% |
$ where to run it
Through an API
api · Volcengine Ark · $0.95/M · $0.47499 in / $2.37494 out per 1M tokens · 256k context · docs.volcengine.com ↗ · price sourceUSD per 1M tokens, base tier (input up to 32K) from models.dev (CNY converted at 6.737012, 2026-08-26); higher tiers for >32K and >128K input and cache excluded. Volcengine pricing page (observed 2026-11-12): doubao-seed-2.0-pro '[0, 32]K: 3.2 / 0.64 / 16.0' CNY. Being retired: Volcengine deprecation notice gives EOM 2026-09-24 and EOS 2026-11-24 14:00 UTC+8 for doubao-seed-2-0-pro-260215 (model list marks it 即将下线); set api to null after EOS.
$ how the score was worked out · ByteDance Seed model page (la… 59%, Epoch AI — internal evaluatio… 40%
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 |
|---|---|---|
| 11 | 59% | |
| 4 | 40% | |
| prior · pull toward the middle | 1% | |
| weighted average | = 70.1 |
$ 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