Magistral Small 2507 needs ~18.4 GB VRAM. RX 9070 16GB has 16.0 GB. With NVFP4 quantization, expect ~19 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
3.6 GB over capacity — needs offload or smaller quantization
Fit status
Too heavy
Decode
14.8 tok/s
TTFT
13120 ms
Safe context
4K
Memory
19.6 GB / 16.0 GB
Offload
20%
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 1.7 GB of extra host RAM just for the offloaded portion, before OS and other tools.
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | A | Very compromised (needs ~1.9 GB host RAM) | 16.8 tok/s | 6276 ms | 4K |
| Coding | F | Too heavy | 14.8 tok/s | 13120 ms | 4K |
| Agentic Coding | F | Too heavy | 11.6 tok/s | 24252 ms | 4K |
| Reasoning | F | Too heavy | 14.8 tok/s | 15505 ms | 4K |
| RAG | F | Too heavy | 11.6 tok/s | 30316 ms | 4K |
How Magistral Small 2507 (24B params) fits at each quantization level on RX 9070 16GB (16.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 9.4 GB | Low | S93 |
Q3_K_SBest for your GPU | 3 | 11.8 GB | Low | S92 |
NVFP4 | 4 | 13.4 GB | Medium | F0 |
Q4_K_M | 4 | 14.6 GB | Medium | F0 |
Q5_K_M | 5 | 17.3 GB | High | F0 |
Q6_K | 6 | 19.7 GB | High | F0 |
Q8_0 | 8 | 25.7 GB | Very High | F0 |
F16 | 16 | 49.2 GB | Maximum | F0 |
Copy-paste commands to run Magistral Small 2507 on your machine.
Run
ollama run magistralOpciones de mejora
Hace que el modelo quepa en el acelerador en lugar de seguir fuera de alcance.
Elimina el offload a memoria del sistema, que suele ser la mayor mejora individual en latencia y throughput.
~$899 MSRP
Hace que el modelo quepa en el acelerador en lugar de seguir fuera de alcance.
Elimina el offload a memoria del sistema, que suele ser la mayor mejora individual en latencia y throughput.
~$999 MSRP
Hace que el modelo quepa en el acelerador en lugar de seguir fuera de alcance.
Elimina el offload a memoria del sistema, que suele ser la mayor mejora individual en latencia y throughput.
~$1,899 MSRP
Yes, RX 9070 16GB can run Magistral Small 2507 at NVFP4 quantization (Very compromised (needs ~1.7 GB host RAM)). The recommended Q4_K_M requires 19.6 GB which exceeds available memory, but at NVFP4 it needs only 18.4 GB. Expected decode speed: 19.2 tok/s.
Magistral Small 2507 (24B parameters) requires approximately 19.6 GB at Q4_K_M quantization. On RX 9070 16GB, it fits at NVFP4 using 18.4 GB.
The recommended quantization is Q4_K_M, but on RX 9070 16GB the best fitting quantization is NVFP4, which uses 18.4 GB.
On RX 9070 16GB, Magistral Small 2507 achieves approximately 19.2 tokens per second decode speed with a time-to-first-token of 10083ms using NVFP4 quantization.
For coding workloads, Magistral Small 2507 on RX 9070 16GB receives a F grade with 14.8 tok/s and 4K context.
On RX 9070 16GB, Magistral Small 2507 can safely use up to 4K tokens of context at NVFP4 quantization. The model's official context limit is 131K, 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/magistral-small-2507-on-rx-9070-16gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>
Preview: