Doubao-Seed-1.6 (thinking): benchmark scores, price and ranking

Doubao-Seed-1.6 (thinking) from ByteDance ranks #379 of 827 LLMs on the LLMs Tiger aggregate benchmark score (48.7), based on 4 benchmark results captured up to 2026-11-18.

Organisation
ByteDance
Type
Frontier model
Open weights
No
Aggregate rank
#379 of 827 (score 48.7)

Relative to other models, Doubao-Seed-1.6 (thinking) is strongest on GPQA Diamond (#135 of 302) and weakest on Chess Puzzles (#80 of 84).

Across the 4 benchmarks where at least ten models are scored, Doubao-Seed-1.6 (thinking) is in the top 10% on 0 and in the top half on 2.

Doubao-Seed-1.6 (thinking) benchmark results

Doubao-Seed-1.6 (thinking) benchmark scores, ranked against all models measured on each benchmark
BenchmarkScoreRankSourceCaptured
GPQA Diamond77.6%#135 of 302Epoch AI Benchmarking Hub2026-11-18
OTIS Mock AIME69.4%#96 of 206Epoch AI Benchmarking Hub2026-11-18
SimpleQA Verified17.4%#73 of 83Epoch AI Benchmarking Hub2026-11-18
Chess Puzzles5.0%#80 of 84Epoch AI Benchmarking Hub2026-11-18

About Doubao-Seed-1.6 (thinking)

Doubao-Seed-1.6 (thinking) is a frontier model from ByteDance. No API price is listed for it yet. We hold 4 benchmark results for it across 2 areas, and none is in the top tenth of its board.

Where Doubao-Seed-1.6 (thinking) stands

On our aggregate score, Doubao-Seed-1.6 (thinking) is #379 of 827 ranked models, around the middle of the table. Its score is 48.7; the scale is centred on 50, so it sits 1.3 points below the typical model. That beats 54% of the ranked field. Among the 7 ranked ByteDance models it is #6. The next one up is Doubao Seed 2.0 Lite (Apr 2026, high) at 51.4. The next one down is seed-2.1-pro-preview at 47.2.

Doubao-Seed-1.6 (thinking), benchmark by benchmark

GPQA Diamond puts Doubao-Seed-1.6 (thinking) at 77.6%, #135 of 302. That beats 55% of the models tested, but 19.3 points short of GPT-6.2 Astra (max) (96.8%). The results either side are claude-sonnet-4-20250514_32K (78.3%) and claude-opus-4-1-20250805_16K (77.3%). It covers expert-level science questions. The board median is 75.1%, so Doubao-Seed-1.6 (thinking) is 2.5 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-11-18.

On OTIS Mock AIME, Doubao-Seed-1.6 (thinking) reaches 69.4%. It ranks #96 of 206, ahead of 53% of the field, and 30.6 points behind Claude Fable 5.1 (max) at 100.0%. The results either side are qwen3-30b-a3b-thinking-2507 (70.3%) and claude-opus-4-1-20250805_27K (68.9%). It covers competition maths (AIME 2024-25 style). The board median is 66.4%, so Doubao-Seed-1.6 (thinking) is 3.1 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-11-18.

SimpleQA Verified is a weak spot. Doubao-Seed-1.6 (thinking) scores 17.4%, #73 of 83, behind 87% of the models measured. The leader, Claude Fable 5.2 (max), reaches 94.0%. It covers short factual questions. The board median is 41.4%, so Doubao-Seed-1.6 (thinking) is 24.0 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-11-18.

Chess Puzzles is a weak spot. Doubao-Seed-1.6 (thinking) scores 5.0%, #80 of 84, behind 94% of the models measured. The leader, Claude Fable 5.2 (max), reaches 79.0%. It covers chess tactics puzzles. The board median is 33.0%, so Doubao-Seed-1.6 (thinking) is 28.0 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-11-18.

Strengths and weaknesses

Reasoning, maths and science: solid, ahead of 54% of the field on average across 2 benchmarks. Its best board here is GPQA Diamond (#135 of 302) and its weakest is OTIS Mock AIME (#96 of 206). General knowledge and chat: weak, ahead of 8% of the field on average across 2 benchmarks. Its best board here is SimpleQA Verified (#73 of 83) and its weakest is Chess Puzzles (#80 of 84).

Reading these numbers

The 4 scores on this page come from 1 source: Epoch AI Benchmarking Hub (4). No official leaderboard supplied any of them, so treat them as third-party measurements. With 4 results the aggregate rank can move a lot when one new score arrives.

Models ranked nearby

More from ByteDance

Compare it on the live leaderboard → · How the aggregate score works