▸ DEVICE UNDER TEST
ARM Mali-G310 MC2 — 2 GB VRAM.
▸ ARM MALI-G310 MC2 SPEC
- BRAND
- Apple
- VRAM
- 2 GB
- BANDWIDTH
- 8.5 GB/s
- FP16 COMPUTE
- 0.06 TFLOPS
- FP32 COMPUTE
- 0.03 TFLOPS
- TDP
- 1 W
- ARCHITECTURE
- Valhall
▸ AI CAPABILITY
57/ 449 models @ Q4
With 2 GB VRAM and 8.5 GB/s bandwidth, this GPU handles models up to 2.13B parameters.
Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~1 tok/s.
§ 01TOP MODELS FOR ARM MALI-G310 MC2
57 FIT · SHOWING 20| MODEL | SIZE | VRAM Q4 | TOK/S | AVG |
|---|---|---|---|---|
| Qwen3-VL 2B Instruct | 2.13B | 1.8 GB | 4 | 18.4 |
| InternVL3 2B | 2.09B | 1.8 GB | 4 | 64.9 |
| Qwen3 1.7B | 2.03B | 1.7 GB | 4 | 30.1 |
| OpenCoder 1.5B Instruct | 1.91B | 1.7 GB | 4 | 47.8 |
| Moondream2 1.9B | 1.9B | 1.6 GB | 4 | 26.2 |
| InternLM2 1B | 1.89B | 1.6 GB | 4 | — |
| Qwen 1.5 1.8B | 1.8B | 1.6 GB | 4 | 19.6 |
| SmolLM2 1.7B | 1.71B | 1.5 GB | 4 | 16.2 |
| Falcon3-1B | 1.67B | 1.5 GB | 5 | 42.0 |
| GPT-2 XL 1.5B | 1.61B | 1.5 GB | 5 | 5.1 |
| stablelm-2-1_6b | 1.6B | 1.5 GB | 5 | 9.5 |
| LFM2-VL 1.6B | 1.6B | 1.5 GB | 5 | 39.7 |
| Falcon-H1 1.5B | 1.55B | 1.4 GB | 5 | 43.8 |
| Qwen2.5-Coder-1.5B | 1.5B | 1.4 GB | 5 | 19.6 |
| Qwen2 Math 1.5B | 1.5B | 1.4 GB | 5 | 19.6 |
| Qwen 2.5 1.5B | 1.5B | 1.4 GB | 5 | 30.2 |
| Yi Coder 1.5B | 1.5B | 1.4 GB | 5 | 14.6 |
| Stella en 1.5B v5 | 1.5B | 1.4 GB | 5 | — |
| Phi-1 1.3B | 1.42B | 1.4 GB | 5 | 7.2 |
| Phi-1.5 1.3B | 1.42B | 1.4 GB | 5 | 7.2 |