NVIDIA RTX 3060 12 GB

The honest floor for NVIDIA CUDA local AI.

A 5-year-old card that still runs Llama 3.1 8B Q4 at 52 tok/s — now at a 5-year-old card's price plus a memory-shortage premium. Verified on Newegg September 22: refurbished 12 GB cards start at $389.99 and the cheapest clearly new US-shipped 12 GB card is $459.99, with marketplace options extending to roughly $660. 12 GB is the minimum VRAM that matters for 8B-class models with useful context, and CUDA still works everywhere.

The decision in five lines

The call: Consider — The honest floor for NVIDIA CUDA local AI.
Best for: Budget entry
Runs well: Qwen 3.5 4B · Qwen 3.5 4B + tight RAG · SANA-0.6B (non-commercial)
Watch out: NVIDIA restarted RTX 3060 12 GB production in June 2026 and we said street prices might soften through Q3. By September 22 the new floor had eased only to $459.99, from $479.99 in August; refurb starts at $389.99. The restart has not restored the old $280–$400 new-card band.
Evidence: Estimated · last verified September 2026

12: GB GDDR6
360: GB/S BANDWIDTH
170: W TDP
~$460: FROM (NEW, SEP 22)

What fits at this tier

Fits 8B dense at Q4 with room for 8–16K context (LocalScore 446, TG 52 tok/s). 14B Q4 technically fits (~8 GB weights) but leaves only ~3 GB for KV cache — short-context work only. 4B-class models fly at 100+ tok/s. No 30B-A3B or 27B dense at this tier.

CODING

Qwen 3.5 4B 4B dense with 262K context; surprisingly coherent for its size.

CHAT / GENERAL

Qwen 3.5 4B 4B dense with 262K context and native multimodal.

DOCS & RETRIEVAL

Qwen 3.5 4B + tight RAG 4B plus tight chunking; keep context windows small.

IMAGE

SANA-0.6B (non-commercial) 0.6B params; <1s per 1024² on a 16GB laptop GPU; weights are NVIDIA NSCL v2 (non-commercial).

AGENTS

Ministral 3 3B Smallest Ministral with reasoning + tool use.

VOICE

Kokoro-82M (Apache 2.0) Community daily driver for English TTS; CPU-real-time at 82M params; v1.0 with 8 languages and 54 voices. No voice cloning.

The call

Buy a refurbished one around $390 if you want a CUDA inference node for 8B chat, document Q&A or learning. New at $460 only makes sense when warranty matters.
Skip the 8 GB variant entirely — it's a different card masquerading under the same name. Also skip if you'll regret not being able to run MoE 30B-A3B — the RTX 5060 Ti 16 GB (~$700 as of August 2026) is where that unlock starts.

Watchouts

NVIDIA restarted RTX 3060 12 GB production in June 2026 and we said street prices might soften through Q3. By September 22 the new floor had eased only to $459.99, from $479.99 in August; refurb starts at $389.99. The restart has not restored the old $280–$400 new-card band.
8 GB variant of the "RTX 3060" exists and is a trap for local AI. Always confirm the 12 GB SKU before buying.
360 GB/s is the bandwidth ceiling — don't expect 3060 to scale past 14B Q4 meaningfully even when weights technically fit.
Used cards from mining rigs often have cosmetic wear and fan wear. Prefer 1-owner gaming pulls; test the card's memory under load on arrival (GPU-Z + a 30-minute stability test).

Local vs cloud at this tier

● LOCAL WINS

Entry-level privacy + local dev loops at 8B scale. Genuinely useful for document summarization, embeddings, local RAG, and chat at the 8B Q4 quality level.

● CLOUD WINS

Cloud wins on quality for anything above 8B. At this tier you're running 2024-era open models; Claude Pro at $20/mo delivers frontier quality at 50+ tok/s with zero hardware risk.

Genuinely worth it as a learning node or a privacy-first 8B daily driver. Break-even vs Claude Pro is ~12 months on cost alone — but the value at this tier is local-itself, not raw quality.

Next step

Load this setup into the planner→