Mistral AI · 119.4B total / 6.5B active · MoE · chat, reasoning, coding, vision
What hardware do you need to run Mistral Small 4 119B-A6B locally?
You need about 78 GB of GPU or unified memory to run Mistral Small 4 119B-A6B at Q4_K_M (4-bit), or 135 GB at Q8_0 (8-bit). 11 of the 18 rigs we track can hold it. The cheapest is the GMKtec EVO-X2 (128 GB) at $3,500, at an estimated ~41 tok/s. The fastest is the PNY NVIDIA RTX PRO 6000 Blackwell at an estimated ~194 tok/s. Prices & stock checked 2026-10-24. Prices change daily and some retailers block our checker, so always confirm on the store page.
Memory needed
| Q4_K_M (4-bit) | Q8_0 (8-bit) | |
|---|---|---|
| Memory needed (est.) | ~78 GB | ~135 GB |
| Weights only | ~72 GB | ~127 GB |
| Read per token | ~3.9 GB | ~6.9 GB |
| Rigs that fit | 11 of 18 | 1 of 18 |
It's a mixture-of-experts model: all 119.4B parameters have to fit in memory, but only 6.5B are read per token, so it runs much faster than a dense model of the same size.
Despite the name, a 119B mixture-of-experts model (128 experts, 4 active, ~6.5B active parameters per token) with image input, 256K context and an Apache 2.0 license, released March 2026. Model card
11 rigs that run Mistral Small 4 119B-A6B
Sorted by today's lowest price. Quant is what we'd use on that machine: Q8_0 when it fits comfortably and still runs 10+ tok/s, otherwise Q4_K_M.
| Rig | Price today | Quant | Memory used | Est. tokens/sec | Buy |
|---|---|---|---|---|---|
| GMKtec EVO-X2128 GB · 256 GB/s | $3,500GMKtec store · 2026-10-24 | Q4_K_M | ~78 of 96 GB | ~41 est. | Check price |
| Minisforum MS-S1 Max128 GB · 256 GB/s | $3,919Minisforum store · 2026-10-24 | Q4_K_M | ~78 of 96 GB | ~41 est. | Check price |
| Beelink GTR9 Pro128 GB · 256 GB/s | $4,349Beelink store · 2026-10-24 | Q4_K_M | ~78 of 96 GB | ~41 est. | Check price |
| Framework Desktop DIY128 GB · 256 GB/s | $4,839Framework store · 2026-10-24 | Q4_K_M | ~78 of 96 GB | ~41 est. | Check price |
| Lenovo ThinkStation PGX128 GB · 273 GB/s | $5,859B&H Photo · 2026-10-24 | Q4_K_M | ~78 of 109 GB | ~45 est. | Check price |
| ASUS Ascent GX10128 GB · 273 GB/s | $5,999B&H Photo · 2026-10-24 | Q4_K_M | ~78 of 109 GB | ~45 est. | Check price |
| NVIDIA DGX Spark128 GB · 273 GB/s | $6,950NVIDIA Marketplace · 2026-10-24 | Q4_K_M | ~78 of 109 GB | ~45 est. | Check price |
| HP ZGX Nano G1n128 GB · 273 GB/s | $7,699B&H Photo · 2026-10-24 | Q4_K_M | ~78 of 109 GB | ~45 est. | Check price |
| Gigabyte AI TOP ATOM128 GB · 273 GB/s | $8,499B&H Photo · 2026-10-24 | Q4_K_M | ~78 of 109 GB | ~45 est. | Check price |
| PNY NVIDIA RTX PRO 6000 Blackwell96 GB · 1,792 GB/s (card only, needs a host PC) | $17,899B&H Photo · 2026-10-24 | Q4_K_M | ~78 of 89 GB | ~194 est. | Check price |
| Mac Studio M5 Ultra256 GB · 1,200 GB/s | No price seen | Q8_0 | ~135 of 192 GB | ~75 est. | Check price |
Tokens/sec are estimates from our bandwidth formula (one user, ~8K context, llama.cpp-class runtime), not benchmarks. GPU prices are for the card only. Prices change daily, so confirm on the store page. How we estimate
Every rig we track, 4-bit and 8-bit
| Machine | Price | Q4_K_M | Q8_0 |
|---|---|---|---|
| Mac mini M5 Pro15-core CPU / 16-core GPU, 24GB | $1,699 | Won't fitneeds ~78 GB | Won't fitneeds ~135 GB |
| GMKtec EVO-X2Ryzen AI Max+ 395, 64GB / 1TB | $2,200 | Won't fitneeds ~78 GB | Won't fitneeds ~135 GB |
| Mac Studio M5 Max18-core CPU / 32-core GPU, 36GB | $2,499 | Won't fitneeds ~78 GB | Won't fitneeds ~135 GB |
| Mac Studio M5 Max18-core CPU / 40-core GPU, 48GB | $3,099 | Won't fitneeds ~78 GB | Won't fitneeds ~135 GB |
| GMKtec EVO-X2Ryzen AI Max+ 395, 128GB / 1TB | $3,500 | Fits~41 tok/s78 of 96 GB | Won't fitneeds ~135 GB |
| Minisforum MS-S1 MaxRyzen AI Max+ 395, 128GB / 2TB | $3,919 | Fits~41 tok/s78 of 96 GB | Won't fitneeds ~135 GB |
| Beelink GTR9 ProRyzen AI Max+ 395, 128GB / 2TB | $4,349 | Fits~41 tok/s78 of 96 GB | Won't fitneeds ~135 GB |
| Framework Desktop DIYRyzen AI Max+ 395, 128GB | $4,839 | Fits~41 tok/s78 of 96 GB | Won't fitneeds ~135 GB |
| Gigabyte GeForce RTX 4090 Windforce V224GB card | $4,899 | Won't fitneeds ~78 GB | Won't fitneeds ~135 GB |
| Mac Studio M5 Ultra30-core CPU / 64-core GPU, 96GB | $5,499 | Won't fitneeds ~78 GB | Won't fitneeds ~135 GB |
| Lenovo ThinkStation PGX1TB | $5,859 | Fits~45 tok/s78 of 109 GB | Won't fitneeds ~135 GB |
| ASUS Ascent GX101TB | $5,999 | Fits~45 tok/s78 of 109 GB | Won't fitneeds ~135 GB |
| NVIDIA DGX SparkFounders Edition, 4TB | $6,950 | Fits~45 tok/s78 of 109 GB | Won't fitneeds ~135 GB |
| HP ZGX Nano G1n4TB | $7,699 | Fits~45 tok/s78 of 109 GB | Won't fitneeds ~135 GB |
| Gigabyte AI TOP ATOM4TB | $8,499 | Fits~45 tok/s78 of 109 GB | Won't fitneeds ~135 GB |
| NVIDIA GeForce RTX 5090 Founders Edition32GB card | $8,500 | Won't fitneeds ~78 GB | Won't fitneeds ~135 GB |
| PNY NVIDIA RTX PRO 6000 Blackwell96GB card | $17,899 | Fits~194 tok/s78 of 89 GB | Won't fitneeds ~135 GB |
| Mac Studio M5 Ultra256GB | — | Fits~96 tok/s78 of 192 GB | Fits~75 tok/s135 of 192 GB |
Fit and tokens/sec are estimates. "Tight fit" means over 90% of usable memory: it loads, but leaves little room for long context.
FAQ
How much memory do I need to run Mistral Small 4 119B-A6B locally?
About 78 GB at Q4_K_M (4-bit) and 135 GB at Q8_0 (8-bit). That's the weights plus runtime overhead and roughly 8K tokens of context for one user. Longer context needs more. These are estimates.
What's the cheapest hardware that runs Mistral Small 4 119B-A6B?
Of the rigs we track, the GMKtec EVO-X2 128 GB at $3,500 (GMKtec store, checked 2026-10-24). We estimate about 41 tokens/sec at 4-bit.
How fast does Mistral Small 4 119B-A6B run locally?
We estimate about 41 to 194 tokens/sec at 4-bit across the rigs that fit, fastest on the PNY NVIDIA RTX PRO 6000 Blackwell.
Can I run Mistral Small 4 119B-A6B on a single gaming GPU like the RTX 4090 or 5090?
Not fully in VRAM. It needs about 78 GB at 4-bit, more than a 24 or 32 GB card. You can offload part of it to system RAM, but it gets much slower. The cheapest rig we track that holds it fully is the GMKtec EVO-X2 128 GB at $3,500.
Can a Mac run Mistral Small 4 119B-A6B?
Yes. The cheapest Mac we track that fits it is the Mac Studio M5 Ultra 256 GB, at an estimated ~96 tok/s. macOS lets the GPU use about 75% of memory by default.
Are these speeds measured or estimated?
The table shows estimates from memory bandwidth, calibrated against 43 public llama.cpp/MLX benchmarks (median error about 5%).