Ryzen AI Max+ 495 Hands On! 160GB of VRAM in a Mini PC, GMKtec EVO X5
This first look evaluates the GMKtec EVO X5 Pro's Ryzen AI Max+ Pro 495, Radeon 865S, and 192 GB of unified memory across hardware inspection, power modes, CPU and GPU benchmarks, large-model inference, and gaming. Its defining advantage over a 128 GB Ryzen 395 system is the ability to allocate 160 GB to the iGPU and run models that do not fit in 96 GB of VRAM.
ETA PRIMEWatchTranscript found
Quick learning frame
Read this before watching.
A model becomes useful when it is wrapped in a harness: tools, state, permissions, memory, routing, and verification.
New playlist item from ETA PRIME; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Skill you build: The ability to evaluate a unified-memory mini PC by separating modest compute-speed gains from the new workloads enabled by a much larger iGPU memory allocation.
Watch for the shift from claim to mechanism. The learning value is the point where the transcript reveals a repeatable action, tool boundary, context move, review habit, or artifact.
Concept diagram
Where this video fits.
01User intent
02Model role
03Tool surface
04State and memory
05Verification loop
06Reusable operating rule
Deep lesson
Turn this video into working knowledge.
2,136 cleaned transcript words reviewed across 600 timed caption segments.
Thesis
Ryzen AI Max+ 495 Hands On! 160GB of VRAM in a Mini PC, GMKtec EVO X5 teaches a practical agent harness move: This first look evaluates the GMKtec EVO X5 Pro's Ryzen AI Max+ Pro 495, Radeon 865S, and 192 GB of unified memory across hardware inspection, power modes, CPU and GPU benchmarks, large-model inference, and gaming. Its defining advantage over a 128 GB Ryzen 395 system is the ability to allocate 160 GB to the iGPU and run models that do not fit in 96 GB of VRAM.
The goal is not to remember the video. The goal is to extract the operating principle, tie it to timestamped evidence, test how far the claim transfers, and make something reusable.
0:59
Unified Memory Headroom
“unified memory. And it is running faster than the old 395. And with this much system memory, we can allocate up to 160 GB to VRAM. Yeah, it's pretty crazy how far they've gone with this. Now, inside...”
The Ryzen AI Max+ Pro 495 combines 16 Zen 5 cores, 32 threads, a Radeon 865S with 40 compute units, and 192 GB of 8,533 MT/s unified memory. Up to 160 GB can be assigned to the iGPU, making memory capacity—not only benchmark speed—the system's standout capability. Build a comparison card listing the 495 system's CPU, iGPU, memory capacity, bandwidth, maximum VRAM allocation, and NPU performance.
4:14
Power Meets Cooling
“that storage. Now, since the memory here, or the RAM, is unified, there's no way to upgrade that, but we can slide the whole shell right off. And this is a pretty beefy dual fan cooling system, as...”
The unit uses a dual-fan cooling system plus a rear SSD fan, while its front button selects quiet, balanced, or performance mode. In performance mode it briefly boosts above 140 W and then sustains about 120 W during the demonstrated stress test. Draw a small test matrix that pairs each power mode with the temperature, sustained wattage, and performance measurements you would collect.
8:23
Capacity Beats Uplift
“in performance. And I also ran 3DMark Time Spy. Our overall score was 11,743. That's an 11.8% increase in performance over that 395. Graphics score on that 495 is sitting at 12,033 versus the 395's 10,809. 11.3% higher,...”
Benchmarks show mostly modest gains over the Ryzen 395, including 3.1% in OpenCL and 6.1% for GPT-OSS, with larger gains in some CPU and 3DMark tests. The more consequential difference is capacity: 160 GB of allocated VRAM runs 320-billion-parameter models that cannot fit on the older system's 96 GB maximum allocation. Separate the reported results into two columns—speed improvements and newly possible workloads—then explain which matters more for large-model inference.
01
User intent
Start with this video's job: This first look evaluates the GMKtec EVO X5 Pro's Ryzen AI Max+ Pro 495, Radeon 865S, and 192 GB of unified memory across hardware inspection, power modes, CPU and GPU benchmarks, large-model inference, and gaming. Its defining advantage over a 128 GB Ryzen 395 system is the ability to allocate 160 GB to the iGPU and run models that do not fit in 96 GB of VRAM. Treat "User intent" as the outcome you are trying to make visible, not a topic label. Anchor it to 0:59, where the video says: “unified memory. And it is running faster than the old 395. And with this much system memory, we can allocate up to 160 GB to VRAM. Yeah, it's pretty crazy how far they've gone with this. Now, inside...”
02
Model role
Use "Model role" to locate the part of the agent harness mechanism the video is demonstrating. Ask what changes in your real setup if this claim is true. Anchor it to 4:14, where the video says: “that storage. Now, since the memory here, or the RAM, is unified, there's no way to upgrade that, but we can slide the whole shell right off. And this is a pretty beefy dual fan cooling system, as...”
03
Tool surface
Turn "Tool surface" into the reusable artifact for this lesson: A one-page agent harness map with tool boundaries, state ownership, and proof signals. This is where watching becomes something you can inspect and reuse.
04
State and memory
Use "State and memory" as the application surface. Decide whether the idea touches a browser flow, a local file, a model choice, a source document, a UI, or a review step.
05
Verification loop
Use "Verification loop" to prove the lesson. The evidence should connect back to the video title, transcript anchors, and a concrete output, not a generic best-practice claim.
06
Reusable operating rule
Use "Reusable operating rule" to carry the idea forward: save the prompt, checklist, diagram, or operating rule that would make the next agent run better.
Example
Source-backed artifact packet
Convert the video into a scoped artifact request that includes the transcript claim, mechanism, acceptance criteria, and proof. The output should be a one-page agent harness map with tool boundaries, state ownership, and proof signals..
Example
Agent harness proof brief
Separate what the speaker claims, what the demo actually proves, and what still needs outside verification before you adopt the agent harness pattern.
Example
Teach-back module
Transform the lesson into a definition, a User intent -> Model role -> Tool surface -> State and memory -> Verification loop -> Reusable operating rule diagram, one misconception, one practice exercise, and a check-for-understanding question.
Do not learn it wrong
Treating the title as the lesson without checking what the transcript actually says.
treating model choice as architecture
ignoring tool permissions
missing verification evidence
Letting the lesson drift into generic agent definitions.
Letting the lesson drift into model leaderboard claims.
Letting the lesson drift into tool list without operating boundaries.
Do not count this as learned until these are true.
01
State the transcript-backed claim in your own words: This first look evaluates the GMKtec EVO X5 Pro's Ryzen AI Max+ Pro 495, Radeon 865S, and 192 GB of unified memory across hardware inspection, power modes, CPU and GPU benchmarks, large-model inference, and gaming. Its defining advantage over a 128 GB Ryzen 395 system is the ability to allocate 160 GB to the iGPU and run models that do not fit in 96 GB of VRAM.
02
Explain the practical stakes without hype: New playlist item from ETA PRIME; queued for transcript-backed review, topic mapping, and a practical learning artifact.
03
Map the idea onto the User intent -> Model role -> Tool surface -> State and memory -> Verification loop -> Reusable operating rule sequence and name the weakest link.
04
Produce the artifact and include the evidence that proves it: A one-page agent harness map with tool boundaries, state ownership, and proof signals.
Put it into practice
Give this grounded prompt to Codex or Claude after watching.
You are helping me turn one specific YouTube video into real, durable learning.
Source video:
- Title: Ryzen AI Max+ 495 Hands On! 160GB of VRAM in a Mini PC, GMKtec EVO X5
- URL: https://www.youtube.com/watch?v=J_35HKz_BZk
- Topic: Agent Architecture
- My current learning frame: Create a purchase-evaluation sheet comparing the 495 and 395 systems on memory fit, inference throughput, CPU and GPU uplift, power behavior, and the specific workloads each can run.
- Why this matters: New playlist item from ETA PRIME; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Transcript anchors from this exact video:
- 0:59 / Evidence 1: "unified memory. And it is running faster than the old 395. And with this much system memory, we can allocate up to 160 GB to VRAM. Yeah, it's pretty crazy how far they've gone with this. Now, inside..."
- 2:32 / Evidence 2: "we've got 16 cores, 32 threads, up to 5.2 GHz. So, we've got a bit of a boost on the core clocks versus the 395. And as for the iGPU, it's the Radeon 865S. 40 compute units, and..."
- 4:14 / Evidence 3: "that storage. Now, since the memory here, or the RAM, is unified, there's no way to upgrade that, but we can slide the whole shell right off. And this is a pretty beefy dual fan cooling system, as..."
- 6:27 / Evidence 4: "performance mode there, for sure. But so far, I mean, this has been a very quick system. Obviously, with that 16-core 32-thread Ryzen APU, we're going to see some great performance on the CPU side. And that GPU,..."
- 8:23 / Evidence 5: "in performance. And I also ran 3DMark Time Spy. Our overall score was 11,743. That's an 11.8% increase in performance over that 395. Graphics score on that 495 is sitting at 12,033 versus the 395's 10,809. 11.3% higher,..."
- 10:02 / Evidence 6: "I've dedicated 160 to VRAM. So, we can run a 320 billion parameter model here. Qwen 3.8 Flash Next, 44 tokens per second. And the GLM 5.3 Flash at 21 tokens per second. And given that the 395..."
- 11:39 / Evidence 7: "Dead Redemption 2. This one works out really well on this 865S. 1440p ultra high mix with no FSR. Looking at an average of around 87 FPS and it looks absolutely amazing like this. If you want to..."
Video-aware target:
- Prompt lane: Agent harness
- Mechanism to extract: Identify what surrounding harness makes the model more useful than chat alone.
- Artifact to produce: A one-page agent harness map with tool boundaries, state ownership, and proof signals.
- Artifact must include: model role; tools; state/memory; permission boundary; verification proof
Your task:
1. Use the transcript anchors above as the primary source packet. If you add outside context, label it clearly as outside context and keep it secondary.
2. Create a source-check table with columns: timestamp, claim, transcript support, what the demo proves, confidence, and what still needs verification.
3. Extract the actual teachable mechanism from the video: Identify what surrounding harness makes the model more useful than chat alone. Do not invent claims that are not supported by the title, lesson frame, or transcript anchors.
4. Build a reusable learning artifact: A one-page agent harness map with tool boundaries, state ownership, and proof signals.
5. Include:
- a plain-English definition of the core idea
- a diagram or structured model using this sequence: User intent -> Model role -> Tool surface -> State and memory -> Verification loop -> Reusable operating rule
- answers to these source questions: What does the video claim the agent can do? | What surrounding system makes that claim plausible? | What proof is shown instead of merely asserted?
- 3 concrete examples that apply the video idea to real agentic work, such as a repo-editing harness; a local research assistant; a recurring refresh agent
- 2 failure modes the video helps prevent, chosen from the transcript evidence and these likely risks: treating model choice as architecture; ignoring tool permissions; missing verification evidence
- a checklist for the next real workflow, focused on: tool boundaries, state ownership, done signal, recovery path
- one practical exercise with a clear done signal: Map one current coding workflow as a harness and mark the first missing proof signal.
6. Add a "learning transfer" section: what changes in my workflow tomorrow if I actually learned this?
7. Add a "source check" section that cites which transcript anchor supports each major takeaway.
Quality bar:
- Make this specific to "Ryzen AI Max+ 495 Hands On! 160GB of VRAM in a Mini PC, GMKtec EVO X5", not a generic Agent Architecture essay.
- Tie each harness element to a transcript anchor that names a tool, state boundary, permission, model behavior, or verification step.
- Prefer operational examples, failure modes, and reusable artifacts over broad definitions.
- Call out uncertainty instead of smoothing over weak evidence.
- Avoid these generic drifts: generic agent definitions; model leaderboard claims; tool list without operating boundaries.
- If evidence is weak or missing, stop and say what transcript segment or timestamp needs review instead of guessing.
- Finish with a concise artifact I could paste into my learning app.
Misconceptions
What to stop believing.
A better model automatically makes a better agent.
The model matters, but harness design determines whether the system can act safely and repeatably.
More tools always help.
Every tool increases surface area. Strong agents have the right tools with clear permissions.
Memory means saving everything.
Useful memory is compressed, curated, and tied to future decisions.
Practice studio
Learning only counts when you make something.
01
Transcript evidence map
Separate what the video actually says from what you already believe about the topic.
3 source-backed takeaways with timestamps, confidence, and a transfer note.02
One useful artifact
Apply the video to a real workflow and produce a one-page agent harness map with tool boundaries, state ownership, and proof signals..
A reusable artifact with a done signal and one verification step.03
Agent harness teach-back card
Explain the agent harness mechanism to someone who has not watched the video yet.
A 90-second explanation, one diagram, one example, and one misconception to avoid.
Recall check
Answer first, then reveal — without rewatching.
What hardware feature lets the EVO X5 Pro allocate 160 GB to its iGPU?
How does the demonstrated performance mode behave under a CPU stress test?
Why is the 495's memory increase more important than its modest AI speed uplift for some models?
Source shelf
Use the video as a doorway, then verify with primary sources.