Claude 3.5 Haiku vs Doubao Seed Code
Wins 0 of 7 areas
—
Wins 6 of 7 areas
Coding · Agents · Reasoning · Images and charts · Long documents · Following instructions
Doubao Seed Code is the stronger all-rounder.
Scores updated · 16 tests both models report · How we compare
Where each one wins
Tests won in each of the seven areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- ReasoningHard problems that need careful thinking02Doubao Seed Code2 of 3 tests · 1 tie
- CodingWriting and fixing software02Doubao Seed Code2 of 2 tests
- AgentsCarrying out multi-step tasks on its own01Doubao Seed Code1 of 1 test
- Images and chartsUnderstanding pictures, charts and video01Doubao Seed Code1 of 1 test
- Long documentsFinding answers in very long texts01Doubao Seed Code1 of 1 test
- Following instructionsDoing exactly what it is asked01Doubao Seed Code1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
Agents, images and charts, long documents and following instructions rest on a single test each.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Doubao Seed Code pulls ahead
- Reasons across sets of long documentsAA-LCR+44.4points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+35.6points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+24.2points ahead
Where Claude 3.5 Haiku pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+38.9points ahead
Every test, side by side
All 16 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingDoubao Seed Code
- Terminal-Bench HardDoubao Seed Code by 24.22.326.5+24.2
- SciCodeDoubao Seed Code by 13.327.440.7+13.3
ReasoningDoubao Seed Code
- GPQA DiamondDoubao Seed Code by 35.640.876.4+35.6
- Humanity's Last ExamDoubao Seed Code by 10.53.614.1+10.5
- CritPttie00.3tie
FactsEven
- AA-Omniscience · Non-hallucinationClaude 3.5 Haiku by 38.958.920+38.9
- AA-Omniscience · AccuracyDoubao Seed Code by 11.713.224.9+11.7
Images and chartsDoubao Seed Code
- MMMU-ProDoubao Seed Code by 22.545.668.1+22.5
Long documentsDoubao Seed Code
- AA-LCRDoubao Seed Code by 44.427.371.7+44.4
Following instructionsDoubao Seed Code
- IFBenchDoubao Seed Code by 8.642.851.4+8.6
Other results5 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- AA Agentic IndexDoubao Seed Code by 34.51.936.4+34.5
- τ²-Bench Telecom (AA run)Doubao Seed Code by 33.624.658.2+33.6
- Artificial Analysis Coding IndexDoubao Seed Code by 15.415.931.3+15.4
- AA-OmniscienceClaude 3.5 Haiku by 12.7-22.5-35.2+12.7
- AA IntelligenceDoubao Seed Code by 88.916.9+8
Questions people ask
Which is better, Claude 3.5 Haiku or Doubao Seed Code?
Doubao Seed Code wins six of the seven areas where both have results: coding, agents, reasoning, images and charts, long documents and following instructions. Claude 3.5 Haiku wins none. They are level on facts.
Which is better for coding?
Doubao Seed Code. It wins 2 of the 2 coding tests both models report; Claude 3.5 Haiku wins none.
How do you compare the two?
We use the 16 benchmark tests both models have published scores on. The verdict counts the 11 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 5 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.