Ten free talking-face engines, one photo, one drawing and one cruel sentence about a man named Bob. What each one did on our RTX 5090, what the licences say, why our lip metric picked the wrong winners, how SoulX-FlashHead Pro got fast enough to go live, and the memory leak we caught before it could eat the PC.
The same sentence, the same photos, four free AI photo editors on one RTX 5090. Which kept the face, which smoothed away the freckles, which re-posed the person, the prompt enhancer that leaks its own reasoning, the one-view prompt that gets a side profile in 12 seconds, and the resolution setting that makes edits nine times slower.
A private ChatGPT for the whole house on a PC with two graphics cards from 2016. Every command, one custom model per person with their own instructions, the two house lines that fix the date and the ‘where do my questions go’ answer, the one setting that stops everyone waiting in line, and the three-day Windows default that killed ours.
A 31-second photoreal film of a consistent AI presenter, made on one RTX 5090 with no cuts: reference pictures given jobs, a locked voice track, three cheap seed previews then one upscaled finish, each clip extending the last. The five ways it broke, the picture that fixed each one, the plate rules, the LoRA retrain, and the bill.
Turn a photograph into an anime character, put that character in six worlds, and make a short film with no dialogue. All of it local, in ComfyUI, on one gaming card. Five things went wrong on the way and all five had the same fix.
One group photograph, three people, and we wanted two of them gone without the third one changing. Five free image editors running on our own card took the job in turn. The first thing that happened was the room came back empty. What fixed it was grammar, and the rest of what we learned is below, with the five workflow files.
Every identity benchmark we can find scores the face. We put one full-length figure through five local image editors with an explicit instruction to keep her proportions, scaled the results so the heads match, and measured. Every engine kept her face. Four of five changed her body. Here is the method, the tool, and what the control test says about who this happens to.
We took a stock Asus TUF gaming laptop with an RTX 3070 and 8 GB of video memory, and benched eight local models overnight across ten tasks. The result that surprised us: a 30 billion parameter model with two thirds of itself off the card beat a 14 billion parameter model that fit better, by nearly four times. File size does not predict speed. All 103 measured rows are free below, failures included.
Twenty generators, eight prompts with invented specifics, one shot each, the prompt on every board. New this round: five Krea 2 builds, Qwen Image 2512, a real cloud column priced in Buzz, and a text ladder scored over four extra seeds. Every board at full size, every prompt in all three styles.
We ran MiniMax H3 for a full day on a single RTX 5090: talking shots driven by our own voice track, a walk and talk, text to speech, a trained voice, and an eight scene short film with its own soundtrack. Every number here came off our machine. So did the mistakes.