Prusa has released firmware 6.9.2 for its Core One INDX toolchanger printers, and a widely shared hands-on review has listed several problems with the machine. The two stories are separate, but anyone considering the INDX for production should read them together.
What happened
According to Prusa's release notes, 6.9.2 is the stable release for the Prusa Core One+ (Gen 2) INDX and the Core One/+ INDX. Prusa's firmware page dates it 7 October 2026 and lists one new feature: filament presets for PVA and BVOH, so those materials can be loaded from the printer. The notes say the firmware is compatible only with those INDX models.
Those presets were meant to arrive earlier. The 6.9.1 notes list them as a feature, then admit they were left out because of a process error. Prusa apologises for that in the 6.9.1 notes. The 6.9.2 notes repeat that the presets were announced in the previous release but not actually included, and say they are now in.
6.9.1 also added a Square Gantry Wizard. It is a standalone wizard placed just before belt tuning. It also runs from inside the belt-tuning wizard after you adjust the screws, to check you have not skewed the gantry.
The review
On 12 October Hackaday wrote up a video by Thomas Sanladerer under the headline "troubled multi-extruder upgrade". Hackaday describes the INDX as an upgrade to the Core One with eight independent extruders. In theory it allows multi-material and multi-colour printing with very little purge waste. Prusa claims 0.013 grams of purge waste per colour or material change.
The problems reported in the video, as Hackaday summarises them:
- The mandatory calibration sequence at the start of each print is designed to work only with Prusa's own filament and its material properties.
- Voxel PLA would not print until the reviewer removed that sequence from the g-code by hand.
- He then found the whole print sequence uses the material presets for Prusa's filament, which led to a failed print.
- Results from manual calibration are not saved even though the display says they are, so the calibration on start-up is effectively compulsory.
- The camera is blocked by the INDX hardware, and looking into the chamber by eye is also heavily obscured.
Hackaday says the issues make the INDX look shaky next to the Bambu Lab H2C, the Snapmaker U1 and Prusa's own XL toolchangers. That is a comparison from one reviewer and a news site, not a test we have run.
What we do not know
The 6.9.2 notes we read do not mention the calibration or third-party filament behaviour, so we cannot say whether Prusa plans to change it. We have not seen a Prusa response to the video. The sources give no price or shipping information for the INDX, so we have not included any. We have not tested the printer ourselves.
What it means for a print farm
This section is our analysis, not reported fact.
The 6.9.2 release itself is small. If you run an INDX and print PVA or BVOH supports, the presets are now on the printer. Otherwise there is little reason to treat it as urgent. The 6.9.1 mix-up is a reminder to check what a release actually contains before you plan work around it. Read the notes, then confirm the feature is on the machine.
The review matters more, and mainly for buyers who print on third-party filament. Most farms buy filament on price and availability, not from one brand. A start-up calibration that depends on one maker's material profile breaks that habit. It also makes any workaround, such as editing g-code, something to repeat on every job. That is a cost in labour and in failed prints. If you cost your jobs properly, a machine that needs hand-editing for your usual spools is more expensive than its purge-waste figure suggests.
The blocked camera is a smaller issue on paper. On a farm it matters more, because remote checking is how one person watches many printers. If you cannot see the chamber, you find out about a bad filament load later.
The low-purge claim is the real attraction for multi-colour work, and nothing we have read disproves it. But it is Prusa's figure, and the review is about whether the machine runs reliably around that figure. If you are thinking of buying, a sensible step is to wait for independent long-run results with the filament you actually use. Another is to try one unit before you commit to several. Compare it against the other toolchanger-style options named above on the jobs you really sell.
