▸ DEVICE UNDER TEST
NVIDIA GeForce GTX 675MX Mac Edition — 1 GB VRAM.
▸ GEFORCE GTX 675MX MAC EDITION SPEC
- BRAND
- NVIDIA
- VRAM
- 1 GB GDDR5
- BANDWIDTH
- 160 GB/s
- FP16 COMPUTE
- 1.9 TFLOPS
- FP32 COMPUTE
- 1.9 TFLOPS
- CUDA CORES
- 1,344
- TDP
- 100 W
- ARCHITECTURE
- Kepler
▸ AI CAPABILITY
22/ 449 models @ Q4
With 1 GB VRAM and 160 GB/s bandwidth, this GPU handles models up to 0.62B parameters.
Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~18 tok/s.
§ 01TOP MODELS FOR GEFORCE GTX 675MX MAC EDITION
22 FIT · SHOWING 20| MODEL | SIZE | VRAM Q4 | TOK/S | AVG |
|---|---|---|---|---|
| Qwen 1.5 0.5B | 0.62B | 0.9 GB | 206 | 9.9 |
| Qwen3-Embedding 0.6B | 0.6B | 0.9 GB | 213 | — |
| Falcon-H1R Tiny 0.6B | 0.6B | 0.9 GB | 213 | — |
| Falcon Perception 0.6B | 0.6B | 0.9 GB | 213 | — |
| BGE-M3 | 0.568B | 0.8 GB | 225 | 63.0 |
| Snowflake Arctic Embed L v2.0 | 0.568B | 0.8 GB | 225 | — |
| Falcon-H1 0.5B | 0.52B | 0.8 GB | 246 | 41.7 |
| Qwen 2.5 0.5B | 0.5B | 0.8 GB | 256 | 19.4 |
| SmolVLM 500M | 0.5B | 0.8 GB | 256 | — |
| GPT-2 Medium 345M | 0.38B | 0.7 GB | 337 | 5.9 |
| SmolLM2 360M | 0.36B | 0.7 GB | 356 | 8.2 |
| LFM2 350M | 0.35B | 0.7 GB | 366 | 46.3 |
| bge-large-en-v1.5 335M | 0.335B | 0.7 GB | 382 | 62.3 |
| mxbai-embed-large-v1 | 0.335B | 0.7 GB | 382 | 64.7 |
| Snowflake Arctic Embed L | 0.335B | 0.7 GB | 382 | 56.0 |
| Snowflake Arctic Embed M v2.0 | 0.305B | 0.7 GB | 420 | — |
| Gemma 3 270M | 0.27B | 0.7 GB | 474 | 12.6 |
| SmolVLM 256M | 0.256B | 0.6 GB | 500 | 28.3 |
| nomic-embed-text-v1.5 100M | 0.14B | 0.6 GB | 914 | 62.3 |
| GPT-2 124M | 0.14B | 0.6 GB | 914 | 6.5 |