Llama 3.3 70B

3 runs · 3 datasets · 1 model

slug: llama-3-3-70b

0.819

Best SPS · opinionsqa

Disaggregated subgroup scorecard. Each card below is one published run for this vendor; expand the question-type and demographic-subgroup sections to see the matrix beneath the headline SPS. Where coverage permits, 95% CI bands accompany the point estimate.

globalopinionqa raw

raw--llama-3.3-70b-instruct--tdefault--tplcurrent--f2748030

0.762 ± 0.030

SPS · 95% CI [0.731, 0.791] · n = 100

Question-type breakdown (7 topics)

Topic	SPS	p_dist	p_rank	p_refuse	N
Health & Science	0.859	0.791	0.927	1.000	2
Technology & Digital Life	0.853	0.803	0.903	0.968	3
Politics & Governance	0.698	0.643	0.754	0.972	22
General Attitudes	0.644	0.666	0.623	1.000	11
International Relations & Security	0.640	0.636	0.644	0.980	50
Economy & Work	0.626	0.661	0.592	0.967	5
Trust & Wellbeing	0.497	0.422	0.572	0.980	7

Not yet measured

This vendor has no demographic-conditioned runs for age, geography, education, or any other subgroup dimension on globalopinionqa. No cells are fabricated — scores appear here only when a conditioned run actually measured them.

How to submit a demographic-conditioned run →

opinionsqa raw

raw--llama-3.3-70b-instruct--tdefault--tplcurrent--6480fa3c

0.819 ± 0.007

SPS · 95% CI [0.812, 0.826] · n = 684

Question-type breakdown (10 topics)

Topic	SPS	p_dist	p_rank	p_refuse	N
Health & Science	0.794	0.753	0.835	0.987	47
Trust & Wellbeing	0.752	0.700	0.803	0.995	25
Media & Information	0.742	0.710	0.775	0.993	63
Economy & Work	0.740	0.705	0.775	0.991	68
Identity & Demographics	0.734	0.659	0.809	0.985	39
Social Values & Religion	0.731	0.686	0.777	0.990	37
General Attitudes	0.726	0.705	0.747	0.989	190
Politics & Governance	0.724	0.661	0.787	0.989	40
International Relations & Security	0.722	0.667	0.777	0.988	149
Technology & Digital Life	0.712	0.681	0.743	0.993	26

Not yet measured

This vendor has no demographic-conditioned runs for age, geography, education, or any other subgroup dimension on opinionsqa. No cells are fabricated — scores appear here only when a conditioned run actually measured them.

How to submit a demographic-conditioned run →

subpop raw

raw--llama-3.3-70b-instruct--tdefault--tplcurrent--670d96d8

0.796 ± 0.016

SPS · 95% CI [0.778, 0.810] · n = 200

Question-type breakdown (9 topics)

Topic	SPS	p_dist	p_rank	p_refuse	N
Trust & Wellbeing	0.875	0.795	0.954	0.988	2
Health & Science	0.807	0.813	0.800	0.994	5
Economy & Work	0.717	0.658	0.776	0.991	17
Social Values & Religion	0.715	0.648	0.783	0.987	36
International Relations & Security	0.711	0.662	0.761	0.991	33
General Attitudes	0.704	0.667	0.741	0.913	37
Politics & Governance	0.704	0.644	0.763	0.983	22
Technology & Digital Life	0.675	0.631	0.720	0.993	47
Identity & Demographics	0.631	0.487	0.774	0.995	1

Not yet measured

This vendor has no demographic-conditioned runs for age, geography, education, or any other subgroup dimension on subpop. No cells are fabricated — scores appear here only when a conditioned run actually measured them.

How to submit a demographic-conditioned run →

No demographic conditioning data has been published for this vendor yet. The question-type matrix above shows topic-level parity; subgroup rows fill in once SynthPanel-style conditioned runs land.

← Back to leaderboard