Skip to content

qwen35-397b-a17b-fp8

sciCORE

legacy model Its test configurations are retired, so it is no longer run. The figures are from its best legacy runs, scored as they ran at the time; it is not ranked.

average±4.1 (95%): 68.0
benchmarksof 12: 12
seconds per inputmean response time: 148.5

Compare with other models Its row in the Matrix All its test runs

By task

Data correction

61.3

Best of 114 models with all its benchmarks: claude-opus-5-5, 82.7.
Document understanding

63.2

Best of 107 models with all its benchmarks: gpt-6-astra, 97.0.
Information extraction

73.7

Best of 107 models with all its benchmarks: grok-4.6, 90.8.
Transcription

89.3

Best of 114 models with all its benchmarks: gemini-4-argon, 96.2.

By benchmark

Benchmark	Score	Rank	Per 1,000 docs	s / input	Best run
Bibliographic Data	45.3±39.2	–	–$0 run	71.71	T1151 · 22 Jun 2026prompt.txt
Blacklist Cards	90.1±3.0	–	–$0 run	45.62	T1158 · 22 Jun 2026prompt.txt
Book Advert XML Repair	24.5±7.4	–	–$0 run	125.35	T1161 · 22 Jun 2026prompt.txt
Business Letters	59.0±6.8	–	–$0 run	255.57	T1152 · 22 Jun 2026prompt.txt
Company Lists	70.3±17.1	–	–$0 run	77.82	T1159 · 22 Jun 2026prompt.txt
Duty Rosters	82.8±3.1	–	–$0 run	394.79	T1341 · 26 Jul 2026prompt.txt
Fraktur Adverts	98.0±0.5	–	–$0 run	236.1	T1155 · 22 Jun 2026prompt_optimized.txt
General Meeting Minutes	86.8±15.8	–	–$0 run	99.46	T1273 · 3 Jul 2026prompt.txt
Library Cards	81.3±1.7	–	–$0 run	95.58	T1156 · 22 Jun 2026prompt.txt
Magazine Pages	8.7±8.2	–	–$0 run	107.46	T1163 · 22 Jun 2026prompt.txt
Medieval Manuscripts	71.6±14.0	–	–$0 run	211.82	T1157 · 22 Jun 2026prompt.txt
Personnel Cards	98.2±0.8	–	–$0 run	60.8	T1162 · 22 Jun 2026prompt.txt

Price

From	Input, $ per million tokens	Output, $ per million tokens	Source
30 Jun 2026	0.0	0.0

Every run