The corpus of surviving Bactrian texts is not enormous, but larger than one might think given how obscure Bactrian culture and the region is considered by much of the rest of the world. Numerous short surviving examples exist in the form of coins, particularly Kushan issue coinage, seals and seal impressions, small inscriptions, but there are also a number of longer manuscripts, chunky fragments of once-larger texts, and more monumental inscriptions.⠀

The major caveat to this is that a lot of these larger texts have only been discovered within the past 20-25 years, so this is a relatively new corpus. The number you'll often see floated around is that the Bactrian language has about 150 surviving documents, but this is a slight misrepresentation in that this actually refers to the large collection of Bactrian texts discovered across the late 90s and 00s, though it does represent the vast majority of substantive Bactrian language documents known to date. That doesn't include either the smaller or highly fragmentary bits and pieces I referred to earlier or the occasional large monumental inscription, the most famous of those arguably being the Rabatak inscription discovered at Surkh Kotal in 1993. ⠀

The other limitation of surviving Bactrian sources is that it heavily weights towards the 2nd-5th centuries AD/CE (there are examples that postdate this, with the last known Bactrian text being from the 9th century AD/CE, but there are a lot fewer of them than the earlier examples at the time of writing), and to my awareness we don't possess any definitive Bactrian or proto-Bactrian examples that predate the Kushan Empire or the 1st century AD/CE. When it comes to pre-Kushan Iranian languages in Bactria we're limited to either a) borrowings/transliterations in Greek or Imperial Aramaic or b) Iranian-isms in Imperial Aramaic inscriptions and documents, which become increasingly common after the end of Achaemenid rule. Neither of those vectors tend to provide insight into early Bactrian or its precursors.⠀

The good caveat, however, is that the Bactrian field tends to evolve rapidly, with more recent discoveries often transforming our understanding in radical ways. The fact that so much of the Bactrian language documents we're aware of were discovered within the past 20 years illustrates that, and there have been small numbers of additional finds since. There is a distinct possibility that, in 10 years, we might come back to this comment and consider it quaintly outdated.⠀

If you're interested in finding the Bactrian language corpus it's not, unfortunately, cohesively collected, but the majority of translated Bactrian documents, i.e the manuscripts+parchments et al, is collected in Bactrian Documents from Northern Afghanistan volumes 1 and 2 by Nicholas Sims-Williams, and there is also an online database that collects a number of the texts untranslated at http://titus.uni-frankfurt.de/texte/etcs/iran/miran/baktr/baktcorp/baktc.htm. I hope that helps answer your question!