Falcon-H1-34B

Falcon-H1-34B places in the top three on 3 benchmarks.

Benchmarks with a score

Ranked first, out of five or more models

In the top three

Technology Innovation Institute May 2025

Open weights Flagship

No Epoch Capabilities Index score yet

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Benchmark	Score	Rank	Source
MT-Bench	9.20	#3 of 23	Model card
IFEval	89.4%	#34 of 126	Model card
TruthfulQA	65.3%	#10 of 28	Model card
Okapi MMLU	77.8%	#3 of 8	Model card
MMLU	84%	#57 of 127	Model card
MMLU-stem	83.6%	#1 of 2	Model card
HellaSwag	81.9%	#29 of 47	Model card
BIG-Bench Hard	70.7%	#44 of 71	Model card
MGSM	76.3%	#33 of 52	Model card
AlpacaEval 2.0	48.3%	#11 of 16	Model card
ARC-Challenge	61%	#33 of 45	Model card
MMLU-Pro	58.7%	#131 of 151	Model card

Benchmark	Score	Rank	Source
AMC	69.4%	#1 of 4	Model card
GPQA Main	41.5%	#4 of 7	Model card
GSM8K	83.6%	#55 of 87	Model card
GPQA Diamond	49.7%	#232 of 284	Model card
MATH-500	83.8%	#57 of 67	Model card
AIME 2024	23.8%	#82 of 87	Model card
AIME 2025	16.7%	#140 of 143	Model card

Benchmark	Score	Rank	Source
MBPP	83.9%	#12 of 66	Model card
HumanEval	87.2%	#31 of 95	Model card
CRUXEval	73.1%	#3 of 9	Model card
MBPP+	71.4%	#9 of 15	Model card
HumanEval+	81.7%	#19 of 27	Model card

Benchmark	Score		Rank	Source
LiveBench	46.3%		#22 of 27	Model card

The benchmarks the lab chose to headline, with the scores it gave.

Every result in the longer documents the lab published with it. 6 of them aren't in the launch post.