M5 Ultra… Apple Wasn’t Messing Around
Credibility score: 69/100 — Mostly Credible. Mixed credibility - some claims are solid, others need verification.
BSmeter analyzed "M5 Ultra… Apple Wasn’t Messing Around" and rated it 69/100 for credibility (a BS score of 31/100 — mostly credible), on 2026-09-22. Its weakest claim — "Claims a 'performance bump' and 'price bump' using 'apparently' twice." — scored 45/100. 29 claims were checked against the video transcript. Scores are produced by BSmeter's AI analysis of the transcript, not independent human verification.
Claims analyzed
Apple's M5 Ultra performance claims — presented as a question of 'real or marketing' — Mostly Credible (75/100)
He's just laying out Apple's claims and then immediately questioning them. That's the setup. 🔥
M5 Ultra memory bandwidth and starting price — straight from the spec sheet — Mostly Credible (75/100)
He's citing the spec sheet for the memory bandwidth and the starting price. No tricks here, just the numbers. 😈
The speaker states the price of the configured M5 Ultra is $14,299. — Mostly Credible (75/100)
He's just laying out the cost of his specific setup. No trickery here, just a number. 😈
Claims a 'performance bump' and 'price bump' using 'apparently' twice. — Mostly Credible (45/100)
He says 'apparently' twice in one breath, trying to sound casual while still making the claim. That's not caution, that's a quiet dodge. 😈
Compares M3 Ultra (47) to M5 Ultra (60.2) Speedometer scores, then pivots to M6 being faster. — Mostly Credible (45/100)
He gives you the M5 score, then immediately undercuts it by saying the M6 is 'much faster.' That's not a comparison, that's a setup and a punchline. 😈
States M3 Ultra took 71.7 seconds for a compilation, M5 Ultra took 52.4. — Mostly Credible (75/100)
He's just giving the raw compilation times for each chip. Straightforward data, no spin. 😈
Claiming M5 Ultra is the fastest he's ever seen for this task. — Mostly Credible (60/100)
His personal 'fastest ever' doesn't make it a universal truth, but it's his experience. 😈
Stating M5 Ultra storage is 'more than two times faster' than M3 Ultra, citing specific benchmarks. — Mostly Credible (75/100)
He's got the numbers right there: 6.5GB/s vs. 14.8GB/s read. The math checks out. 😈
Asserting 'two times faster' is a 'conservative label' and that the M5 Ultra is 'actually faster than that'. — Mostly Credible (45/100)
He's saying Apple's marketing is 'conservative' based on his own tests. Bold claim for a mortal. 😈
M5 Ultra's GPU cores have dedicated neural accelerators, unlike M3 Ultra. — Mostly Credible (75/100)
He's laying out the technical specs for the M5 Ultra — sounds like a genuine architectural upgrade. 😈
Llama CPP shows 'has tensor = true' on M5 Ultra, 'false' on M3 Ultra. — Mostly Credible (75/100)
He's showing a direct, verifiable software output difference between the chips. That's called evidence, mortals. 😈
Incogn sponsor read for data removal service. — Mostly Credible (50/100)
Ah, the old 'protect your data' pitch. This is a commercial, plain and simple. 😈
Incogni ad: "Make your personal information harder to find with Incogni." — Mostly Credible (50/100)
Ah, the classic mid-roll pitch. They're selling you peace of mind, which is always a hot commodity. 🔥
CPU benchmark shows lower memory bandwidth than Apple's claim, then explains why. — Mostly Credible (75/100)
He's showing the numbers, then immediately explaining the discrepancy. That's called being upfront, mortals. 😈
GPU benchmark results for M3 and M5 Ultra, aligning with Apple's claims. — Mostly Credible (75/100)
He's showing the numbers, then immediately explaining the discrepancy. That's called being upfront, mortals. 😈
Asks if token generation will be 1.4x faster, setting up the next test. — Mostly Credible (75/100)
A direct question, setting up the next test. No tricks here, just curiosity. 😈
Deepseek V4 Flash token generation: M3 Ultra at 37 tokens/sec, M5 Ultra at 53 tokens/sec, 1.5x faster. — Mostly Credible (75/100)
He's showing the actual performance numbers and the speedup. The data speaks for itself. 😈
Compares M3 Ultra to M5 Ultra performance, notes Apple's higher claim. — Mostly Credible (75/100)
He's giving numbers and then immediately flagging the discrepancy with Apple's own claims. Straightforward reporting. 😈
Updates M3 Ultra performance and compares it to M5 Ultra on a specific model. — Mostly Credible (75/100)
More numbers, more specific models, and he's even re-tested it himself. This is what mortals call 'evidence.' 😈
Demonstrates M3 Ultra vs. M5 Ultra performance with large context window. — Mostly Credible (75/100)
He's showing a real-world test with massive context. The numbers are right there, no smoke and mirrors. 😈
Comparing M3 Ultra to M5 Ultra token generation — straightforward numbers. — Mostly Credible (75/100)
He's just laying out the numbers, M3 vs M5. No tricks, just data. 😈
Claiming Apple's 'four times faster' prompt processing is accurate, based on his own test. — Mostly Credible (75/100)
He's verifying Apple's claim with his own test. The numbers match up. Fine. 😈
Qualifying the 'four times faster' claim by showing it depends on prompt length. — Mostly Credible (45/100)
He's admitting the 'four times faster' is conditional, but only after selling the big number. That's a classic volume game. 🚩
Downplaying efficiency gains by stating tokens per watt only increased by 12%. — Mostly Credible (75/100)
He's just giving the efficiency numbers. No spin, just the raw data. 😈
Comparing M3 Ultra's 200 watts to M5 Ultra's 400+ watts under heavy GPU load. — Mostly Credible (75/100)
He's showing the GPU history chart right there, the numbers are on display. No trickery, just raw data. 🔥
Claiming the M5 Ultra is 'definitely much much more audible' and can be heard through his microphone. — Mostly Credible (60/100)
He's telling you what he hears and what his mic picks up. It's his experience, not a scientific claim. 👂
Stating the M5 Ultra's 'bright orange' color on the display indicates it's heating up 'quite a bit more'. — Mostly Credible (75/100)
The visual evidence is right there on the screen, showing the M5 Ultra's temperature reading in a more intense color. It's a direct observation. 🌡️
Stating token generation is 1.5x faster and prompt processing 2-4x faster, which coding agents will 'feel'. — Mostly Credible (75/100)
He's giving specific performance metrics and a direct benefit. It's a clear statement of observed improvement. 🚀
Previewing a test where the M5 Ultra outputs 134 tokens/second against the M3 Ultra's 75 tokens/second. — Mostly Credible (75/100)
He's giving a direct comparison from a specific test, showing the numbers side-by-side. It's a clear benchmark. 📊
See the full analysis with sources and timestamps →