# Best local LLM for 192GB RAM or VRAM

URL: https://insidethemachine.ai/best/local-llm-for-192gb-ram
Data: models 2026-10-15, machines 2026-10-15. Source of scores: Artificial Analysis Intelligence Index (AA), LMArena. Not our own tests yet.
Usable memory assumed: 144 GB (75% of 192). Rule: min_memory_gb = weights_gb_4bit × 1.15 + 2. weights_gb_4bit = Q4_K_M file (Ollama, else HF GGUF) → MLX 4-bit (HF) → Ollama MLX/NVFP4 tag → Ollama default tag → params_total_b × 0.6 (flagged weights_estimated). +15% ≈ KV cache at ~8–16K context; +2GB runtime/process overhead. Long contexts (128K+) need much more.
Models that fit: 94 of 124.

## Top picks

| model | maker | AA index | needs (4-bit) | needs (8-bit) | run |
|---|---|---|---|---|---|
| Qwen3.8-Flash-Next (125B-A6B) | Alibaba Qwen | 39.8 | 140 GB | 219.3 GB | `ollama run qwen3.8-flash-next:125b-a6b` |
| Qwen3.8 27B | Alibaba Qwen | 33.7 | 22.7 GB | 36.5 GB | `ollama run qwen3.8:27b` |
| Nemotron 3.5 Super VL 120B-A12B | NVIDIA | 26.5 | 105.5 GB | 154.9 GB | `ollama run nemotron-3.5-super:120b-a12b` |
| K2 Horizon MoVA 36B-A4B | IFM (MBZUAI) | 25.3 | 28.1 GB | 47.8 GB | MLX only |
| Qwen3.5 27B | Alibaba Qwen | 22.9 | 21.5 GB | 36.5 GB | `ollama run qwen3.5:27b` |
| Qwen3.6 27B | Alibaba Qwen | 21.4 | 21.5 GB | 36.5 GB | `ollama run qwen3.6:27b` |

## By job

- Coding: Qwen3.8-Flash-Next (125B-A6B) (140 GB, Terminal-Bench 2.1 86.1%) `ollama run qwen3.8-flash-next:125b-a6b`
- Agents (tool use): Qwen3.8-Flash-Next (125B-A6B) (140 GB) `ollama run qwen3.8-flash-next:125b-a6b`
- Vision (reads images): Qwen3.8-Flash-Next (125B-A6B) (140 GB)

## New at 192 GB (did not fit in 128 GB)

- Qwen3.8-Flash-Next (125B-A6B) (140 GB, AA 39.8)
- Nemotron 3.5 Super VL 120B-A12B (105.5 GB, AA 26.5)
- Nemotron 3 Super 120B-A12B (102 GB, AA 12.8)
- Laguna S 2.1 118B-A8B (112.4 GB, AA unscored)

## Machines with 192 GB

| machine | type | usable GB | models that fit | best fit | from (approx) |
|---|---|---|---|---|---|
| GMKtec EVO-X5 Pro (Ryzen AI Max+ PRO 495 "Gorgon Halo") | AMD mini PC | 159.9 | 94 | Qwen3.8-Flash-Next (125B-A6B) | $6,799 (2026-10-15) |
| Minisforum MS-S1 MAX-P495 (Ryzen AI Max+ PRO 495 "Gorgon Halo") | AMD mini PC | 159.9 | 94 | Qwen3.8-Flash-Next (125B-A6B) | $7,399 (2026-10-16) |
| AMD Ryzen AI Halo Developer Platform (Ryzen AI Max+ PRO 495, 192GB) | AMD mini PC | 159.9 | 94 | Qwen3.8-Flash-Next (125B-A6B) | unknown |

## Questions

**What is the best local LLM for 192 GB of RAM?**
Qwen3.8-Flash-Next (125B-A6B) by Alibaba Qwen. It scores 39.8 on the Artificial Analysis Intelligence Index and needs about 140 GB at 4-bit. That is about 69% of Claude Opus 5.5 (57.6), the top paid model. Data as of 2026-10-15.

**How many local models fit in 192 GB?**
94 of the 124 text models in our data fit. We assume about 144 GB is usable for the model (75% of 192 GB), and a model needs its 4-bit file size × 1.15 + 2 GB.

**What is the biggest model that runs in 192 GB?**
Qwen3.8-Flash-Next (125B-A6B), 180B (6B active) parameters, needs about 140 GB at 4-bit. Bigger is not always smarter: it is also the top pick.

**What is the best coding model for 192 GB?**
Qwen3.8-Flash-Next (125B-A6B) (needs 140 GB, Terminal-Bench 2.1 86.1%). Run it in a coding app like OpenCode or Cline. See https://insidethemachine.ai/best/local-coding-model.

**Does 192 GB of VRAM work the same as 192 GB of RAM?**
Close. On a Mac the memory is shared, and about 75% goes to the model. On a PC the graphics card's VRAM is what counts; we estimate VRAM minus 1 GB is usable, so a 192 GB card fits a little more than 192 GB of Mac memory. Models that spill out of VRAM into normal RAM run much slower.

More: https://insidethemachine.ai/models?tier=192 · https://insidethemachine.ai/machines?tier=192 · https://insidethemachine.ai/best