Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Seed 1.6 is clearly ahead on the aggregate, 65 to 49. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Seed 1.6's sharpest advantage is in coding, where it averages 42.4 against 19.3. The single biggest benchmark swing on the page is SWE-bench Verified, 46 to 17.
Seed 1.6 is the reasoning model in the pair, while Claude 3 Haiku is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. Seed 1.6 gives you the larger context window at 256K, compared with 200K for Claude 3 Haiku.
Pick Seed 1.6 if you want the stronger benchmark profile. Claude 3 Haiku only becomes the better choice if you would rather avoid the extra latency and token burn of a reasoning model.
Seed 1.6
62.3
Claude 3 Haiku
44
Seed 1.6
42.4
Claude 3 Haiku
19.3
Seed 1.6
79.6
Claude 3 Haiku
68.7
Seed 1.6
74.5
Claude 3 Haiku
59.6
Seed 1.6
56.4
Claude 3 Haiku
42.8
Seed 1.6
87
Claude 3 Haiku
76
Seed 1.6
83.4
Claude 3 Haiku
71.1
Seed 1.6
75.9
Claude 3 Haiku
62.2
Seed 1.6 is ahead overall, 65 to 49. The biggest single separator in this matchup is SWE-bench Verified, where the scores are 46 and 17.
Seed 1.6 has the edge for knowledge tasks in this comparison, averaging 56.4 versus 42.8. Inside this category, FrontierScience is the benchmark that creates the most daylight between them.
Seed 1.6 has the edge for coding in this comparison, averaging 42.4 versus 19.3. Inside this category, SWE-bench Verified is the benchmark that creates the most daylight between them.
Seed 1.6 has the edge for math in this comparison, averaging 75.9 versus 62.2. Inside this category, AIME 2023 is the benchmark that creates the most daylight between them.
Seed 1.6 has the edge for reasoning in this comparison, averaging 74.5 versus 59.6. Inside this category, MuSR is the benchmark that creates the most daylight between them.
Seed 1.6 has the edge for agentic tasks in this comparison, averaging 62.3 versus 44. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.
Seed 1.6 has the edge for multimodal and grounded tasks in this comparison, averaging 79.6 versus 68.7. Inside this category, OfficeQA Pro is the benchmark that creates the most daylight between them.
Seed 1.6 has the edge for instruction following in this comparison, averaging 87 versus 76. Inside this category, IFEval is the benchmark that creates the most daylight between them.
Seed 1.6 has the edge for multilingual tasks in this comparison, averaging 83.4 versus 71.1. Inside this category, MGSM is the benchmark that creates the most daylight between them.
Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.
Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.