Describe your build — any GPUs, CPU, RAM, OS, runtime, use case. We'll compute effective VRAM honestly, recommend a runtime, and tell you which models fit comfortably, which are borderline, and which aren't practical.
Total VRAM ≠ pooled VRAM. We never sum VRAM unless the silicon truly pools (Apple unified memory). We always explain why effective is lower than total.
Add GPUs, set CPU/RAM/OS, optionally pick a runtime + use case. URL updates as you change fields — share a build by copying the URL.
Single NVIDIA GeForce RTX 3090 — 24 GB VRAM minus ~1.5 GB runtime overhead = ~22 GB usable for weights + KV cache + activations. The 8% headroom we reserve covers the typical OS/driver footprint and gives KV-cache room for an 8K-32K context.
Best engine for this topology + skill level + use case.
183 models considered. Categorized by headroom at the recommended quant + a sensible context for your use case.
| Model | Params | Quant | VRAM est. | Context | Note |
|---|---|---|---|---|---|
| OLMo 2 13B | 13B | Q4_K_M | 17.4 GB | 4,096 | Fits cleanly at Q4_K_M + 4,096 ctx with 24% headroom. |
| Stable LM 2 12B | 12B | Q4_K_M | 16.5 GB | 4,096 | Fits cleanly at Q4_K_M + 4,096 ctx with 28% headroom. |
| Tulu 3 8B | 8B | Q4_K_M | 19.1 GB | 8,192 | Fits cleanly at Q4_K_M + 8,192 ctx with 17% headroom. |
| Molmo 7B-D | 8B | Q4_K_M | 13 GB | 4,096 | Comfortable fit with 44% headroom — room to extend context or run alongside other workloads. |
| Llama 3.1 Nemotron Nano 8B | 8B | Q4_K_M | 19.1 GB | 8,192 | Fits cleanly at Q4_K_M + 8,192 ctx with 17% headroom. |
| Granite 3.0 8B Instruct | 8B | Q4_K_M | 13 GB | 4,096 | Comfortable fit with 44% headroom — room to extend context or run alongside other workloads. |
| DeepSeek R1 Distill Llama 8B | 8B | Q4_K_M | 19.1 GB | 8,192 | Fits cleanly at Q4_K_M + 8,192 ctx with 17% headroom. |
| MiniCPM-V 2.6 8B | 8B | Q4_K_M | 19.1 GB | 8,192 | Fits cleanly at Q4_K_M + 8,192 ctx with 17% headroom. |
| OpenCoder 8B | 8B | Q4_K_M | 19.1 GB | 8,192 | Fits cleanly at Q4_K_M + 8,192 ctx with 17% headroom. |
| Granite 3.2 8B | 8B | Q4_K_M | 19.1 GB | 8,192 | Fits cleanly at Q4_K_M + 8,192 ctx with 17% headroom. |
| InternLM 3 8B | 8B | Q4_K_M | 19.1 GB | 8,192 | Fits cleanly at Q4_K_M + 8,192 ctx with 17% headroom. |
| Granite 3.3 8B | 8B | Q4_K_M | 19.1 GB | 8,192 | Fits cleanly at Q4_K_M + 8,192 ctx with 17% headroom. |
| Ministral 8B Instruct | 8B | Q4_K_M | 19.1 GB | 8,192 | Fits cleanly at Q4_K_M + 8,192 ctx with 17% headroom. |
| MiniCPM-V 3 8B | 8B | Q4_K_M | 19.1 GB | 8,192 | Fits cleanly at Q4_K_M + 8,192 ctx with 17% headroom. |
| Aya 23 8B | 8B | Q4_K_M | 19.1 GB | 8,192 | Fits cleanly at Q4_K_M + 8,192 ctx with 17% headroom. |
| Llama 3.3 8B Instruct | 8B | Q4_K_M | 19.1 GB | 8,192 | Fits cleanly at Q4_K_M + 8,192 ctx with 17% headroom. |
| EXAONE 3.5 8B | 8B | Q4_K_M | 18.8 GB | 8,192 | Fits cleanly at Q4_K_M + 8,192 ctx with 18% headroom. |
| CodeGemma 7B | 7B | Q4_K_M | 17.9 GB | 8,192 | Fits cleanly at Q4_K_M + 8,192 ctx with 22% headroom. |
| Falcon Mamba 7B | 7B | Q4_K_M | 17.9 GB | 8,192 | Fits cleanly at Q4_K_M + 8,192 ctx with 22% headroom. |
| Mistral 7B Instruct v0.3 | 7B | Q5_K_M | 18.5 GB | 8,192 | Fits cleanly at Q5_K_M + 8,192 ctx with 19% headroom. |
| Qwen 2.5 7B Instruct | 7B | Q8_0 | 18.3 GB | 8,192 | Fits cleanly at Q8_0 + 8,192 ctx with 21% headroom. |
| Qwen 2.5-VL 7B | 7B | Q4_K_M | 17.9 GB | 8,192 | Fits cleanly at Q4_K_M + 8,192 ctx with 22% headroom. |
| Codestral Mamba 7B | 7B | Q4_K_M | 17.9 GB | 8,192 | Fits cleanly at Q4_K_M + 8,192 ctx with 22% headroom. |
| Qwen 3 7B | 7B | Q4_K_M | 17.9 GB | 8,192 | Fits cleanly at Q4_K_M + 8,192 ctx with 22% headroom. |
| Model | Params | Quant | VRAM est. | Context | Note |
|---|---|---|---|---|---|
| DeepSeek MoE 16B Base | 16B | Q4_K_M | 20 GB | 4,096 | Tight fit at Q4_K_M — only 13% headroom. KV cache for longer context will OOM. Cap context tighter or drop one quant level. |
| Llama 3.2 11B Vision Instruct | 11B | Q4_K_M | 22.5 GB | 8,192 | Tight fit at Q4_K_M — only 2% headroom. KV cache for longer context will OOM. Cap context tighter or drop one quant level. |
| Llama 3.2 11B Vision | 11B | Q4_K_M | 22.5 GB | 8,192 | Tight fit at Q4_K_M — only 2% headroom. KV cache for longer context will OOM. Cap context tighter or drop one quant level. |
| Falcon 3 10B | 10B | Q4_K_M | 21.3 GB | 8,192 | Tight fit at Q4_K_M — only 7% headroom. KV cache for longer context will OOM. Cap context tighter or drop one quant level. |
| Gemma 2 9B Instruct | 9B | Q4_K_M | 20.2 GB | 8,192 | Tight fit at Q4_K_M — only 12% headroom. KV cache for longer context will OOM. Cap context tighter or drop one quant level. |
| Yi Coder 9B | 9B | Q4_K_M | 20.2 GB | 8,192 | Tight fit at Q4_K_M — only 12% headroom. KV cache for longer context will OOM. Cap context tighter or drop one quant level. |
| Nemotron 3 Nano 9B | 9B | Q4_K_M | 20.2 GB | 8,192 | Tight fit at Q4_K_M — only 12% headroom. KV cache for longer context will OOM. Cap context tighter or drop one quant level. |
| GLM-4 9B | 9B | Q4_K_M | 20.2 GB | 8,192 | Tight fit at Q4_K_M — only 12% headroom. KV cache for longer context will OOM. Cap context tighter or drop one quant level. |
| Hermes 3 Llama 3.1 8B | 8B | Q8_0 | 22.9 GB | 8,192 | Tight fit at Q8_0 — only 0% headroom. KV cache for longer context will OOM. Cap context tighter or drop one quant level. |
| Qwen 3 8B | 8B | Q8_0 | 22.9 GB | 8,192 | Tight fit at Q8_0 — only 0% headroom. KV cache for longer context will OOM. Cap context tighter or drop one quant level. |
| Llama 3.1 8B Instruct | 8B | Q8_0 | 20 GB | 8,192 | Tight fit at Q8_0 — only 13% headroom. KV cache for longer context will OOM. Cap context tighter or drop one quant level. |
| DeepSeek R1 Distill Qwen 7B | 7B | Q8_0 | 21.3 GB | 8,192 | Tight fit at Q8_0 — only 7% headroom. KV cache for longer context will OOM. Cap context tighter or drop one quant level. |
| Qwen 2.5 Coder 7B Instruct | 7B | Q6_K | 19.6 GB | 8,192 | Tight fit at Q6_K — only 15% headroom. KV cache for longer context will OOM. Cap context tighter or drop one quant level. |
| Model | Params | Quant | VRAM est. | Context | Note |
|---|---|---|---|---|---|
| NV-Embed v2 | 8B | FP16 | 30.4 GB | 8,192 | ~30.4 GB needed at FP16 + 8,192 ctx — overshoots effective VRAM by 32%. Drop quant or move to a larger build. |
| Qwen 3 Embedding 8B | 8B | FP16 | 30.8 GB | 8,192 | ~30.8 GB needed at FP16 + 8,192 ctx — overshoots effective VRAM by 34%. Drop quant or move to a larger build. |
| PaliGemma 2 10B | 10B | BF16 | 36 GB | 8,192 | ~36.0 GB needed at BF16 + 8,192 ctx — overshoots effective VRAM by 56%. Drop quant or move to a larger build. |
| Pixtral 12B | 12B | Q4_K_M | 23.6 GB | 8,192 | ~23.6 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 3%. Drop quant or move to a larger build. |
| Gemma 3 12B | 12B | Q4_K_M | 23.6 GB | 8,192 | ~23.6 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 3%. Drop quant or move to a larger build. |
| Mistral Nemo 12B Instruct | 12B | Q4_K_M | 23.6 GB | 8,192 | ~23.6 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 3%. Drop quant or move to a larger build. |
| Baichuan 4 13B | 13B | Q4_K_M | 24.7 GB | 8,192 | ~24.7 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 8%. Drop quant or move to a larger build. |
| GLM-4V 9B | 14B | Q4_K_M | 25.8 GB | 8,192 | ~25.8 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 12%. Drop quant or move to a larger build. |
| DeepSeek R1 Distill Qwen 14B | 14B | Q4_K_M | 25.9 GB | 8,192 | ~25.9 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 12%. Drop quant or move to a larger build. |
| Qwen 2.5 14B Instruct | 14B | Q4_K_M | 25.9 GB | 8,192 | ~25.9 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 12%. Drop quant or move to a larger build. |
| Phi-4 Multimodal | 14B | Q4_K_M | 25.9 GB | 8,192 | ~25.9 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 12%. Drop quant or move to a larger build. |
| Qwen 2.5 Coder 14B Instruct | 14B | Q4_K_M | 25.9 GB | 8,192 | ~25.9 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 12%. Drop quant or move to a larger build. |
| Phi-4 14B | 14B | Q4_K_M | 25.9 GB | 8,192 | ~25.9 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 12%. Drop quant or move to a larger build. |
| Phi-4 Reasoning 14B | 14B | Q4_K_M | 25.9 GB | 8,192 | ~25.9 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 12%. Drop quant or move to a larger build. |
| Qwen 3 14B | 14B | Q4_K_M | 25.9 GB | 8,192 | ~25.9 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 12%. Drop quant or move to a larger build. |
| StarCoder 2 15B | 15B | Q4_K_M | 27 GB | 8,192 | ~27.0 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 17%. Drop quant or move to a larger build. |
NVLink vs PCIe, tensor- vs pipeline-parallel, mixed-card honesty.
Curated multi-GPU / cluster setups with effective-VRAM math.
OS + runtime install commands for your stack.
Runtime × OS × hardware support truth table.
If you're sizing a fresh AI build (not just a card to drop into an existing system), the build-budget walkthroughs cover the whole BOM honestly: AI PC build under $1,000 or AI PC build under $2,000 cover the realistic 2026 budget tiers.
Vertical-fit shopping? AI PC for students covers the budget + portability tradeoffs; AI PC for developers covers the coding workflow specifics; AI PC for small business covers the document-RAG / always-on machine.
Form-factor first? See best laptop for local AI, best Mac for local AI, best mini PC for local AI, or best used GPU for local AI.