Llama 4 Maverick 17B 128E needs ~219.0 GB VRAM. AMD Instinct MI300X 192GB has 192.0 GB. With Q3_K_S quantization, expect ~36 tok/s.
Operating mode
Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.
Current mode
Balanced
Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.
Select quantization to explore
75.0 GB over capacity — needs offload or smaller quantization
Fit status
Too heavy
Decode
20.5 tok/s
TTFT
9435 ms
Safe context
4K
Memory
267.0 GB / 192.0 GB
Offload
30%
It fits through host-memory offload, and offload is the main reason performance drops.
CPU or host-memory offload is active
About 10% of the working set spills out of accelerator memory, which usually hurts latency and sustained decode throughput.
Very little memory headroom
You can run the model, but there is not much room left for longer context, bigger batches, extra apps, or future model updates.
Remove offload with more accelerator memory
Prioritize a GPU or unified-memory tier that fits the whole model natively. Removing offload usually helps more than small compute gains.
Buy headroom, not only minimum fit
A slightly larger memory tier gives you safer context growth and makes the recommendation more future-proof.
Increase host RAM if you keep offloading
This setup may need roughly 24.2 GB of extra host RAM just for the offloaded portion, before OS and other tools.
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | F | Too heavy | 20.8 tok/s | 5087 ms | 4K |
| Coding | F | Too heavy | 20.5 tok/s | 9435 ms | 4K |
| Agentic Coding | F | Too heavy | 20.1 tok/s | 14043 ms | 4K |
| Reasoning | F | Too heavy | 20.5 tok/s | 11151 ms | 4K |
| RAG | F | Too heavy | 20.1 tok/s | 17553 ms | 4K |
How Llama 4 Maverick 17B 128E (400B params) fits at each quantization level on AMD Instinct MI300X 192GB (192.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 156.0 GB | Low | F0 |
Q3_K_S | 3 | 196.0 GB | Low | F0 |
NVFP4 | 4 | 224.0 GB | Medium | F0 |
Q4_K_M | 4 | 244.0 GB | Medium | F0 |
Q5_K_M | 5 | 288.0 GB | High | F0 |
Q6_K | 6 | 328.0 GB | High | F0 |
Q8_0 | 8 | 428.0 GB | Very High | F0 |
F16 | 16 | 820.0 GB | Maximum | F0 |
Copy-paste commands to run Llama 4 Maverick 17B 128E on your machine.
Run
lms load Llama-4-Maverick-17B-128E-Instruct && lms server startUpgrade-Optionen
Makes the model fit on the accelerator instead of staying completely out of reach.
Removes host-memory offload, which is usually the single biggest latency and throughput win.
ca. $8,000 MSRP
Makes the model fit on the accelerator instead of staying completely out of reach.
Raises estimated decode speed by about 85%.
ca. $20,000 MSRP
Yes, AMD Instinct MI300X 192GB can run Llama 4 Maverick 17B 128E at Q3_K_S quantization (Very compromised (needs ~24.2 GB host RAM)). The recommended Q4_K_M requires 267.0 GB which exceeds available memory, but at Q3_K_S it needs only 219.0 GB. Expected decode speed: 36.1 tok/s.
Llama 4 Maverick 17B 128E (400B parameters) requires approximately 267.0 GB at Q4_K_M quantization. On AMD Instinct MI300X 192GB, it fits at Q3_K_S using 219.0 GB.
The recommended quantization is Q4_K_M, but on AMD Instinct MI300X 192GB the best fitting quantization is Q3_K_S, which uses 219.0 GB.
On AMD Instinct MI300X 192GB, Llama 4 Maverick 17B 128E achieves approximately 36.1 tokens per second decode speed with a time-to-first-token of 5370ms using Q3_K_S quantization.
For coding workloads, Llama 4 Maverick 17B 128E on AMD Instinct MI300X 192GB receives a F grade with 20.5 tok/s and 4K context.
On AMD Instinct MI300X 192GB, Llama 4 Maverick 17B 128E can safely use up to 4K tokens of context at Q3_K_S quantization. The model's official context limit is 1.0M, but available memory constrains the safe maximum.
Remove offload with more accelerator memory. Prioritize a GPU or unified-memory tier that fits the whole model natively. Removing offload usually helps more than small compute gains.
Paste this snippet into any page to show a live fit card.
<iframe src="https://willitrunai.com/embed/llama-4-maverick-17b-128e-on-instinct-mi300x-192gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>
Preview: