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.
Calculations follow the RunLocalAI Will-It-Run Framework: effective VRAM, model working set, runtime constraints, fit tiers, and measured-vs-estimated evidence labels.
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.
4× NVIDIA H100 SXM = 320 GB total VRAM, but without NVLink, cross-card bandwidth is PCIe-bound (~32 GB/s vs NVLink ~112 GB/s). With tensor-parallelism, each card holds ~1/4 of the model weights and replicates activations + KV cache. After 15% TP overhead, effective model capacity is ~265 GB. Largest single tensor on one card is ~78 GB.
Publicly inspectable measured rows for the selected hardware slug(s). Exact measured rows calibrate the fit table instead of leaving it as pure VRAM estimation.
No publicly inspectable benchmark rows are attached to this exact hardware yet. The engine will still calculate fit and runtime, but speed rows will remain estimated.
Best engine for this topology + skill level + use case.
Tensor-parallel across NVLink/PCIe — works on every recent consumer + datacenter pair. AWQ-INT4 + 70B fits dual 3090 / dual 4090 cleanly.
Single-stream king. EXL2 4.0bpw + 70B fits dual 3090 with NVLink and beats vLLM on solo-user throughput.
Layer-split via --tensor-split is the experimentation-friendly path. Worse throughput than vLLM but easier to debug.
315 models considered. Categorized by headroom at the recommended quant + a sensible context for your use case.
| Model | Params | Quant | VRAM est. | Context | Evidence | Note |
|---|---|---|---|---|---|---|
| Kimi K1.5 | 200B | AWQ-INT4 | 220.8 GB | 8,192 | No measured row yet | Fits cleanly at AWQ-INT4 + 8,192 ctx with 17% headroom. |
| GLM-5 Pro | 144B | AWQ-INT4 | 158.1 GB | 8,192 | No measured row yet | Comfortable fit with 40% headroom — room to extend context or run alongside other workloads. |
| Mixtral 8x22B Instruct | 141B | Q4_K_M | 158.7 GB | 8,192 | No measured row yet | Comfortable fit with 40% headroom — room to extend context or run alongside other workloads. |
| WizardLM-2 8x22B | 141B | Q4_K_M | 158.7 GB | 8,192 | No measured row yet | Comfortable fit with 40% headroom — room to extend context or run alongside other workloads. |
| DBRX Base | 132B | Q4_K_M | 144.8 GB | 8,192 | No measured row yet | Comfortable fit with 45% headroom — room to extend context or run alongside other workloads. |
| DBRX Instruct | 132B | AWQ-INT4 | 144.8 GB | 8,192 | No measured row yet | Comfortable fit with 45% headroom — room to extend context or run alongside other workloads. |
| Mistral Large 2 (123B) | 123B | Q4_K_M | 138.2 GB | 8,192 | No measured row yet | Comfortable fit with 48% headroom — room to extend context or run alongside other workloads. |
| Nemotron 3 Super (120B-A12B) | 120B | Q4_K_M | 135.6 GB | 8,192 | No measured row yet | Comfortable fit with 49% headroom — room to extend context or run alongside other workloads. |
| Llama 4 Scout | 109B | Q5_K_M | 136.4 GB | 8,192 | No measured row yet | Comfortable fit with 49% headroom — room to extend context or run alongside other workloads. |
| Sarvam 105B | 105B | Q4_K_M | 113.2 GB | 8,192 | No measured row yet | Comfortable fit with 57% headroom — room to extend context or run alongside other workloads. |
| Sarvam 105B FP8 | 105B | Q4_K_M | 113.2 GB | 8,192 | No measured row yet | Comfortable fit with 57% headroom — room to extend context or run alongside other workloads. |
| Command R+ 104B | 104B | Q4_K_M | 115 GB | 8,192 | No measured row yet | Comfortable fit with 57% headroom — room to extend context or run alongside other workloads. |
| Command R+ (Aug 2024) | 104B | AWQ-INT4 | 115 GB | 8,192 | No measured row yet | Comfortable fit with 57% headroom — room to extend context or run alongside other workloads. |
| Llama 3.2 90B Vision | 90B | AWQ-INT4 | 99.6 GB | 8,192 | No measured row yet | Comfortable fit with 62% headroom — room to extend context or run alongside other workloads. |
| Llama 3.2 90B Vision Instruct | 90B | Q4_K_M | 98.6 GB | 8,192 | No measured row yet | Comfortable fit with 63% headroom — room to extend context or run alongside other workloads. |
| InternVL 2.5 78B | 78B | Q4_K_M | 86.3 GB | 8,192 | No measured row yet | Comfortable fit with 67% headroom — room to extend context or run alongside other workloads. |
| Qwen 2.5 Math 72B | 72B | Q4_K_M | 61.1 GB | 4,096 | No measured row yet | Comfortable fit with 77% headroom — room to extend context or run alongside other workloads. |
| Qwen 2.5 72B Instruct | 72B | Q5_K_M | 87.5 GB | 8,192 | No measured row yet | Comfortable fit with 67% headroom — room to extend context or run alongside other workloads. |
| Molmo 72B | 72B | Q4_K_M | 61.1 GB | 4,096 | No measured row yet | Comfortable fit with 77% headroom — room to extend context or run alongside other workloads. |
| Qwen 2.5-VL 72B | 72B | AWQ-INT4 | 80.1 GB | 8,192 | No measured row yet | Comfortable fit with 70% headroom — room to extend context or run alongside other workloads. |
| Dolphin 3 Llama 3.3 70B | 70B | AWQ-INT4 | 77 GB | 8,192 | No measured row yet | Comfortable fit with 71% headroom — room to extend context or run alongside other workloads. |
| DeepSeek R1 Distill Llama 70B | 70B | Q5_K_M | 84.4 GB | 8,192 | No measured row yet | Comfortable fit with 68% headroom — room to extend context or run alongside other workloads. |
| Llama 3.1 70B Instruct | 70B | Q5_K_M | 84.4 GB | 8,192 | No measured row yet | Comfortable fit with 68% headroom — room to extend context or run alongside other workloads. |
| Tulu 3 70B | 70B | Q4_K_M | 77 GB | 8,192 | No measured row yet | Comfortable fit with 71% headroom — room to extend context or run alongside other workloads. |
| Model | Params | Quant | VRAM est. | Context | Evidence | Note |
|---|---|---|---|---|---|---|
| DeepSeek Coder V2 236B | 236B | Q4_K_M | 258.7 GB | 8,192 | No measured row yet | Tight fit at Q4_K_M — only 2% headroom. KV cache for longer context will OOM. Cap context tighter or drop one quant level. |
| DeepSeek V2.5 236B | 236B | Q4_K_M | 258.7 GB | 8,192 | No measured row yet | Tight fit at Q4_K_M — only 2% headroom. KV cache for longer context will OOM. Cap context tighter or drop one quant level. |
| K-EXAONE 236B A23B | 236B | Q4_K_M | 254.3 GB | 8,192 | No measured row yet | Tight fit at Q4_K_M — only 4% headroom. KV cache for longer context will OOM. Cap context tighter or drop one quant level. |
| GLM-5 | 200B | Q4_K_M | 226 GB | 8,192 | No measured row yet | Tight fit at Q4_K_M — only 15% headroom. KV cache for longer context will OOM. Cap context tighter or drop one quant level. |
| Model | Params | Quant | VRAM est. | Context | Evidence | Note |
|---|---|---|---|---|---|---|
| Qwen 3 235B-A22B | 235B | Q4_K_M | 266.6 GB | 8,192 | No measured row yet | ~266.6 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 1%. Drop quant or move to a larger build. |
| Llama 3.1 Nemotron Ultra 253B | 253B | Q4_K_M | 277.7 GB | 8,192 | No measured row yet | ~277.7 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 5%. Drop quant or move to a larger build. |
| DeepSeek V4 Flash (284B MoE) | 284B | Q4_K_M | 312.1 GB | 8,192 | No measured row yet | ~312.1 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 18%. Drop quant or move to a larger build. |
| Hunyuan Large 389B MoE | 389B | Q4_K_M | 425.5 GB | 8,192 | No measured row yet | ~425.5 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 61%. Drop quant or move to a larger build. |
| Qwen 3.5 235B-A17B (MoE) | 397B | Q4_K_M | 435.8 GB | 8,192 | No measured row yet | ~435.8 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 64%. Drop quant or move to a larger build. |
| Jamba 1.5 Large | 398B | Q4_K_M | 440.5 GB | 8,192 | No measured row yet | ~440.5 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 66%. Drop quant or move to a larger build. |
| Llama 4 Maverick | 400B | Q4_K_M | 452 GB | 8,192 | No measured row yet | ~452.0 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 71%. Drop quant or move to a larger build. |
| Llama 4 405B | 405B | AWQ-INT4 | 444 GB | 8,192 | No measured row yet | ~444.0 GB needed at AWQ-INT4 + 8,192 ctx — overshoots effective VRAM by 68%. Drop quant or move to a larger build. |
| DeepSeek V3 (671B MoE) | 671B | Q4_K_M | 734.5 GB | 8,192 | No measured row yet | ~734.5 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 177%. Drop quant or move to a larger build. |
| DeepSeek R1 (671B reasoning) | 671B | Q4_K_M | 734.5 GB | 8,192 | No measured row yet | ~734.5 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 177%. Drop quant or move to a larger build. |
| Mistral Medium 3.5 (675B MoE) | 675B | Q4_K_M | 744.9 GB | 8,192 | No measured row yet | ~744.9 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 181%. Drop quant or move to a larger build. |
| DeepSeek V4 | 745B | AWQ-INT4 | 818.8 GB | 8,192 | No measured row yet | ~818.8 GB needed at AWQ-INT4 + 8,192 ctx — overshoots effective VRAM by 209%. Drop quant or move to a larger build. |
| Kimi K2.6 | 1000B | Q4_K_M | 1130 GB | 8,192 | No measured row yet | ~1130.0 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 326%. Drop quant or move to a larger build. |
| Ring-2.6-1T | 1000B | Q3_K_M | 1011.9 GB | 8,192 | No measured row yet | ~1011.9 GB needed at Q3_K_M + 8,192 ctx — overshoots effective VRAM by 282%. Drop quant or move to a larger build. |
| Step-3 | 1000B | AWQ-INT4 | 1093.3 GB | 8,192 | No measured row yet | ~1093.3 GB needed at AWQ-INT4 + 8,192 ctx — overshoots effective VRAM by 313%. Drop quant or move to a larger build. |
| DeepSeek V4 Pro (1.6T MoE) | 1600B | Q4_K_M | 1766 GB | 8,192 | No measured row yet | ~1766.0 GB needed at Q4_K_M + 8,192 ctx — overshoots effective VRAM by 566%. 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.