What can you run with 128GB of RAM?
With 128GB RAM you can run 19 of the 30 open-source models we track — up to about 199B parameters at Q4 quantization. The most popular that fits is Qwen3.5-9B, needing 6.8 GB and generating around ~62 tok/s on typical hardware of this size.
Typical machines: Mac Studio M3 Ultra · Threadripper workstation · Multi-GPU server. After the operating system takes its share, about 133 GB is free for a model.
Having the memory is not the same as having the speed
128GB RAM decides whether a model loads. How fast it answers is decided by memory bandwidth — how quickly your machine can read the model, which it must do once for every word it writes. Machines with 128GB RAM span 90–819 GB/s, so the fastest is about 9.1× quicker than the slowest with identical capacity. The table quotes a Mac M4 Max; pick yours below to see the difference. Why bandwidth and not the processor →
Models that run on 128GB RAM in 2026
| Model | Size | Memory | Est. speed | Feels like |
|---|---|---|---|---|
| Qwen3.5-9BAlibaba / Qwen | 9.7B | 6.8 GBRuns well | ~62 tok/s | Fast |
| gpt-oss-20bOpenAI · MoE | 22B | 13 GBRuns well | ~153 tok/s4.3B active | Fast |
| Qwen3.5-4BAlibaba / Qwen | 4.7B | 3.8 GBRuns well | ~129 tok/s | Fast |
| gpt-oss-120bOpenAI · MoE | 120B | 66 GBRuns well | ~28 tok/s24B active | Fast |
| Qwen3.5-0.8BAlibaba / Qwen | 873M | 1.3 GBRuns well | ~666 tok/s | Fast |
| Qwen3.5-27BAlibaba / Qwen | 28B | 19 GBRuns well | ~21 tok/s | Comfortable |
| Qwen3.5-35B-A3BAlibaba / Qwen · MoE | 36B | 25 GBRuns well | ~81 tok/s7.2B active | Fast |
| Qwen3.5-2BAlibaba / Qwen | 2.3B | 2.0 GBRuns well | ~277 tok/s | Fast |
| GLM-4.7-FlashZhipu AI | 31B | 22 GBRuns well | ~19 tok/s | Comfortable |
| Qwen3.5-122B-A10BAlibaba / Qwen · MoE | 125B | 82 GBRuns well | ~23 tok/s25B active | Comfortable |
| Olmo-3-7B-InstructAi2 | 7.3B | 7.2 GBRuns well | ~79 tok/s | Fast |
| Qwen3-Next-80B-A3B-InstructAlibaba / Qwen · MoE | 81B | 52 GBRuns well | ~37 tok/s16B active | Fast |
| GLM-4.6V-FlashZhipu AI | 10B | 7.0 GBRuns well | ~58 tok/s | Fast |
| ERNIE-4.5-VL-28B-A3B-PTBaidu AI · MoE | 29B | 20 GBRuns well | ~20 tok/s | Comfortable |
| Mistral-Small-4-119B-2603Mistral AI | 119B | 79 GBRuns well | ~4.8 tok/s | Slow |
| Step-3.5-FlashStepFun | 199B | 85 GBRuns well | ~4.9 tok/s | Slow |
| Olmo-3.1-32B-InstructAi2 | 32B | 23 GBRuns well | ~18 tok/s | Comfortable |
| ERNIE-4.5-21B-A3B-PTBaidu AI · MoE | 22B | 15 GBRuns well | ~26 tok/s | Fast |
| ERNIE-4.5-0.3B-PTBaidu AI | 361M | 853 MBRuns well | ~1475 tok/s | Fast |
All figures at Q4_K_M quantization, 4K context. “Memory” includes the model, its context cache and runtime overhead. Speed is calculated, not benchmarked: 546 GB/s × 0.65 ÷ bytes read per token, assuming a Mac M4 Max. Scale it by your own machine's bandwidth from the list above. Mixture-of-experts models read only their active parameters per token, which is why some large ones outrun smaller dense models.
Image generators that also fit
These make pictures rather than text. Their memory use scales with the resolution you generate at, not conversation length, so treat these figures as a floor rather than a ceiling.
What you'd need more memory for
The next models up, and what they ask for.