Fix misleading single vs dual card comparison

Single card table was showing A- for 35B-A3B which undercut the case
for a second card. Restructured to lead with 16GB (start here) at
honest B+ ratings, then show what 32GB unlocks: dense 27B/32B models
that physically can't fit on 16GB and are genuinely A- quality.
The upgrade isn't about 262K context - it's about accessing better
models (27B dense > 35B MoE with 3B active params).

https://claude.ai/code/session_01PtYTPherSJaxDEVPgF6Nxu

docs/gpu-setup-research.md

@@ -218,32 +218,39 @@ A 3-slot bridge leaves an air gap for proper cooling.
 this works well in rack airflow. If using a 2U server, measure clearance for the NVLink
 bridge sitting on top of the cards. A 4U chassis gives the most room.
 
-### What Runs on 32GB Unified (2x RTX 5000 + NVLink)
+### What Runs on 16GB (Single RTX 5000 — Start Here)
 
-| Model | Arch | Quant | Weights | Context | Total VRAM | Quality |
+| Model | Quant | Context | Quality | Notes |
-|-------|------|-------|---------|---------|------------|---------|
+|-------|-------|---------|---------|-------|
-| **Qwen3.5-35B-A3B** | **MoE (3B active)** | **Q4_K_M** | **~12GB** | **262K** | **~25GB** | **B+ to A-** |
+| **Qwen3.5-35B-A3B** | Q4_K_L | ~64-128K | **B+** | MoE, 3B active. Good but VRAM is tight — context may be lower |
-| Qwen3.5-27B | Dense | Q4_K_M | ~17GB | 128K+ | ~25GB | A- |
+| Qwen3.5-9B | Q8 | 128K+ | B+ | Fits comfortably, high quant |
-| Qwen3-Coder-Next (80B/3B active) | MoE | Q4 | ~20GB | 128K | ~28GB | A |
+| Qwen3.5-4B | Q8 | 262K | B | Tiny model, long context |
-| Qwen2.5-Coder-14B | Dense | Q4_K_M | ~10GB | 128K | ~22GB | B+ |
+| Qwen2.5-Coder-7B | Q8 | 128K | B | Solid for simple tasks |
-| Qwen2.5-Coder-14B | Dense | Q8 | ~16GB | 64K | ~28GB | A- |
+| Qwen2.5-Coder-14B | Q4_K_M | 16-32K | B+ | Tight fit, limited context |
-| Qwen2.5-Coder-32B | Dense | Q4_K_M | ~20GB | 16-24K | ~28GB | A- |
 
-**The sweet spot: Qwen3.5-35B-A3B at Q4_K_M with 262K context.** This is a 35B parameter
+A single card is a solid start — B+ coding with decent context. But 16GB is the ceiling.
-model with only 3B active at inference (MoE). The Gated Delta Network architecture slashes
+You can't run bigger dense models, can't use higher quantization, and context is squeezed.
-KV cache memory. The entire model + full 262K context fits in ~25GB — well within 32GB.
 
-### What Runs on 16GB (Single RTX 5000 — Phase 1)
+### What the Second Card + NVLink Unlocks (32GB)
 
-| Model | Quant | Context | Quality |
+The second card doesn't just double context — it opens models that **don't fit on 16GB at all:**
-|-------|-------|---------|---------|
-| **Qwen3.5-35B-A3B** | Q4_K_L | ~64-128K | **A-** |
-| Qwen3.5-9B | Q8 | 128K+ | B+ |
-| Qwen3.5-4B | Q8 | 262K | B |
-| Qwen2.5-Coder-7B | Q8 | 128K | B |
-| Qwen2.5-Coder-14B | Q4_K_M | 16-32K | B+ |
 
-Even a single card can run the Qwen3.5-35B-A3B MoE model — just with a smaller context window.
+| Model | Arch | Quant | Weights | Context | Total VRAM | Quality | **Why it needs 32GB** |
+|-------|------|-------|---------|---------|------------|---------|----------------------|
+| **Qwen3.5-27B** | **Dense** | **Q4_K_M** | **~17GB** | **128K+** | **~25GB** | **A-** | **17GB weights won't fit on 16GB** |
+| **Qwen2.5-Coder-32B** | **Dense** | **Q4_K_M** | **~20GB** | **16-24K** | **~28GB** | **A-** | **20GB weights won't fit on 16GB** |
+| Qwen2.5-Coder-14B | Dense | **Q8** | ~16GB | 64K | ~28GB | A- | Q8 quant = better output, needs 16GB for weights alone |
+| Qwen3-Coder-Next (80B) | MoE | Q4 | ~20GB | 128K | ~28GB | A | 20GB weights won't fit on 16GB |
+| Qwen3.5-35B-A3B | MoE (3B active) | Q4_K_M | ~12GB | 262K | ~25GB | B+ | Fits on 1 card at reduced context, but 32GB = full 262K + headroom |
+
+**The real upgrade isn't 262K context — it's access to dense 27B/32B models that are
+genuinely A- quality.** The 35B-A3B MoE runs on both setups, but its 3B active params
+limit quality. The Qwen3.5-27B dense model uses all 27B params on every token — that's
+the quality jump. And its 17GB of weights physically can't fit on a single 16GB card.
+
+Think of it this way:
+- **1 card**: B+ coding (MoE or small dense models, squeezed context)
+- **2 cards**: **A- coding** (full dense 27B/32B models, comfortable context, higher quant options)
 
 ### Estimated Inference Speed