Best Local AI Models for Every VRAM Tier (4GB to 32GB+)

You are helping me turn one specific YouTube video into real, durable learning.

Source video:
- Title: Best Local AI Models for Every VRAM Tier (4GB to 32GB+)
- URL: https://www.youtube.com/watch?v=JQfA5WzRKN8
- Topic: Creative Automation
- My current learning frame: Match a model to your actual VRAM tier using this video's quantization math, then before buying a bigger card, run that same model through a second agent harness or scaffold to see how much performance you can unlock for free.
- Why this matters: New playlist item from Cloud Codes; queued for transcript-backed review, topic mapping, and a practical learning artifact.

Transcript anchors from this exact video:
- 0:00 / Evidence 1: "Last month, something shifted in the Steam Hardware Survey that changes what you can run on your own machine. 16 GB of video memory became the most common graphics card configuration that people actually own. First time in..."
- 3:56 / Evidence 2: "reasoning dense rather than merely large, and it holds up on structured output and tool calls, which is most of what a small local model actually gets asked to do. It also scores six on the artificial analysis..."
- 8:05 / Evidence 3: "It is downloaded 12.4 million times a month from Hugging Face. Apache 2.0. 262,000 tokens of context natively, extensible past a million, and it carries a vision encoder, so it reads images. And then there is the row..."
- 9:41 / Evidence 4: "result. And notice which benchmarks are not on that card at all. Human Eval, the coding test the older guide still list. MTBench, the conversational one. Not reported, not on the QN cards, not on the Gemma cards,..."
- 12:27 / Evidence 5: "the smaller model on the benchmark that matters most for agents by the same company in the same year. And the 27B is the one that fits on a card you can buy in a shop. It is..."
- 13:58 / Evidence 6: "experts, 10 firing at a time. 262,000 tokens of context. Apache 2.0. Half a million downloads a month. 70.6 on SWE-Bench Verified. 44.3 on SWE-Bench Pro. 36.2 on Terminal Bench. Those are serious agentic coding numbers from an..."
- 16:17 / Evidence 7: "long context retrieval, a benchmark for knowing what you do not know. It is a deliberately unkind test, which is what makes it useful. So, the gap is 25 index points between the best model you can fit..."

Video-aware target:
- Prompt lane: Local model/runtime
- Mechanism to extract: Identify why the local setup works or fails for this specific agent task, not whether local models are generally good.
- Artifact to produce: A local model/runtime fit sheet with hardware constraints, model choice, endpoint setup, task benchmark, and fallback rule.
- Artifact must include: hardware; runtime; model/quantization; endpoint; agent integration; benchmark/fallback

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 why the local setup works or fails for this specific agent task, not whether local models are generally good. Do not invent claims that are not supported by the title, lesson frame, or transcript anchors.
4. Build a reusable learning artifact: A local model/runtime fit sheet with hardware constraints, model choice, endpoint setup, task benchmark, and fallback rule.
5. Include:
   - a plain-English definition of the core idea
   - a diagram or structured model using this sequence: Task -> Hardware -> Model/quantization -> Runtime endpoint -> Agent tool loop -> Benchmark task -> Fallback
   - answers to these source questions: What machine/runtime is shown? | What task exposes the model limit? | What setup change improves the loop?
   - 3 concrete examples that apply the video idea to real agentic work, such as Ollama or LM Studio coding endpoint; MLX Apple Silicon runner; DGX-backed Hermes session
   - 2 failure modes the video helps prevent, chosen from the transcript evidence and these likely risks: using a local model like ChatGPT; ignoring latency/context limits; no benchmark task
   - a checklist for the next real workflow, focused on: task fit, runtime setup, latency/context, tool loop, fallback
   - one practical exercise with a clear done signal: Choose one real coding task and specify the pass/fail benchmark for a local model.
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 "Best Local AI Models for Every VRAM Tier (4GB to 32GB+)", not a generic Creative Automation 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: local-model ideology; hardware specs without workflow fit; benchmarks unrelated to the actual task.
- 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.