MacBook Pro M4 32GBBudget pick
32 GB Unified (+12)
B
The model fits in shared memory, but shared-memory bandwidth is now the real limiter.2 tok/s decode
~$799 MSRP
Nomic Embed Text v1.5 needs ~4.0 GB VRAM. RX 7900 XT 20GB has 20.0 GB. With F16 quantization, expect ~2 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
Fit status
Runs well
Decode
2.0 tok/s
TTFT
96800 ms
Safe context
8K
Memory
4.0 GB / 20.0 GB
This model fits, but memory bandwidth is the part holding decode speed back.
Throughput will feel slow
Estimated decode speed is only 2.0 tok/s, so this is more of a technical fit than a comfortable daily-driver setup.
Prioritize bandwidth, not only capacity
If this workload feels slow, the next useful step is often a GPU tier with materially faster memory bandwidth rather than only a small bump in capacity.
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | B | Runs well | 2.0 tok/s | 52800 ms | 8K |
| Coding | B | Runs well | 2.0 tok/s | 96800 ms | 8K |
| Agentic Coding | B | Runs well | 2.0 tok/s | 140800 ms | 8K |
| Reasoning | B | Runs well | 2.0 tok/s | 114400 ms | 8K |
| RAG | B | Runs well | 2.0 tok/s | 176000 ms | 8K |
Inference speed
Estimated decode speed (tokens/sec) for Nomic Embed Text v1.5 at F16 across popular GPUs and Apple Silicon, using the fastest local runtime per device. Fastest is RTX 5090 32GB at ~2 tok/s. Speed is memory-bandwidth bound, so cards that fit the whole model in VRAM run far faster than ones that offload to system RAM.
| GPU / Mac | Memory | Quant | Speed (tok/s) | Fits? |
|---|---|---|---|---|
 RTX 5090 32GB | 32 GB | F16 | 2.0 | Fits |
| RTX 4090 24GB | 24 GB | F16 | 2.0 | Fits |
| RTX 4080 Super 16GB | 16 GB | F16 | 2.0 |
How Nomic Embed Text v1.5 (0.13699999451637268B params) fits at each quantization level on RX 7900 XT 20GB (20.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 0.1 GB | Low | A77 |
Q3_K_S | 3 | 0.1 GB | Low | A77 |
NVFP4 | 4 |
Copy-paste commands to run Nomic Embed Text v1.5 on your machine.
Run
ollama run nomic-embed-textUpgrade options
~$799 MSRP
~$1,099 MSRP
Yes, RX 7900 XT 20GB can run Nomic Embed Text v1.5 with a B grade (Runs well). Expected decode speed: 2.0 tok/s.
Nomic Embed Text v1.5 (0.13699999451637268B parameters) requires approximately 4.0 GB of memory with F16 quantization.
The recommended quantization for Nomic Embed Text v1.5 is F16, which balances quality and memory efficiency.
On RX 7900 XT 20GB, Nomic Embed Text v1.5 achieves approximately 2.0 tokens per second decode speed with a time-to-first-token of 96800ms using F16 quantization.
For coding workloads, Nomic Embed Text v1.5 on RX 7900 XT 20GB receives a B grade with 2.0 tok/s and 8K context.
On RX 7900 XT 20GB, Nomic Embed Text v1.5 can safely use up to 8K tokens of context. The model's official context limit is 8K, but available memory constrains the safe maximum.
Paste this snippet into any page to show a live fit card.
<iframe src="https://willitrunai.com/embed/nomic-embed-text-v1.5-on-rx-7900-xt-20gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>
Preview:
| RTX 3090 24GB | 24 GB | F16 | 2.0 | Fits |
| RTX 4070 12GB | 12 GB | F16 | 2.0 | Fits |
| RTX 3060 12GB | 12 GB | F16 | 2.0 | Fits |
| RTX 4060 8GB | 8 GB | F16 | 2.0 | Fits |
RX 7900 XTX 24GB | 24 GB | F16 | 2.0 | Fits |
MacBook Pro M4 Max 128GB | 128 GB | F16 | 2.0 | Fits |
Mac Studio M3 Ultra 256GB | 256 GB | F16 | 2.0 | Fits |
Mac Studio M2 Ultra 128GB | 128 GB | F16 | 2.0 | Fits |
Mac Studio M1 Ultra 128GB | 128 GB | F16 | 2.0 | Fits |
MacBook Pro M4 Max 64GB | 64 GB | F16 | 2.0 | Fits |
MacBook Pro M3 Max 64GB | 64 GB | F16 | 2.0 | Fits |
MacBook Pro M1 Max 64GB | 64 GB | F16 | 2.0 | Fits |
MacBook Pro M4 Pro 48GB | 48 GB | F16 | 2.0 | Fits |
Estimates for single-stream decoding at F16; real tokens/sec varies with prompt length, context, batch size, and runtime build. Prompt processing (prefill) is faster than the decode figures shown here.
| Medium |
| A77 |
Q4_K_M | 4 | 0.1 GB | Medium | A77 |
Q5_K_M | 5 | 0.1 GB | High | A77 |
Q6_K | 6 | 0.1 GB | High | A77 |
Q8_0 | 8 | 0.1 GB | Very High | A77 |
F16Best for your GPU | 16 | 0.3 GB | Maximum | A77 |
Prioritize bandwidth, not only capacity. If this workload feels slow, the next useful step is often a GPU tier with materially faster memory bandwidth rather than only a small bump in capacity.