Opt 125M — Hardware Requirements & GPU Compatibility
Chat
Opt 125M is a 125M-parameter open language model from optimum-intel-internal-testing. It supports a context window of up to 2,048 tokens. At FP16 it needs about 0.28 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 8 GB (FP16 · 2K)Best on 24 GB (FP16 · ~2340 tok/s · 2K)Macs from 32 GB (~300 tok/s)
Other
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Opt 125M
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
vllm serve optimum-intel-internal-testing/opt-125m --max-model-len 2048vLLM repo with its own quantisation
Download: Weights on Hugging Face
Other ways
each opens step-by-stepRun in cloud
Fits on RTX 3060 12GB (11 GB headroom) · FP16
- Generation speed
- ~836 tok/s
- generation speed
- Cost per 1M output tokens
- $0.024
- per 1M output tokens
How Much VRAM Does Opt 125M Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 0.3 GB | — | 16-bit · full — Full half-precision — baseline for inference | Fits: 2K context · ~2340 tok/s |
Selected FP16
Full half-precision0.3GBat 2K context
- Weights
- 0.1 GB
- KV cache
- 0.0 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 23.7 GB free · ~2340 tok/s
≥50 tok/s needs ≥22 GB/s · RTX 5090 ~4160, RTX 3090 Ti ~2340
Which Hardware Can Run Opt 125M?
FP16 · 0.3 GBOpt 125M (FP16) requires 0.3 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 1+ GB is recommended. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 176 devices with unified memory can run Opt 125M, including NVIDIA DGX H100, NVIDIA DGX A100 640GB.
≤ 11 GB
≤ 11 GB tier →fits · 7.7 GB free · up to 2K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~1765Simulate output on RTX 3080RTX 2080 Ti~1430Simulate output on RTX 2080 TiRTX 3070 Ti~1412Simulate output on RTX 3070 TiM3*8 GB~256Simulate output on iMac 24" M3 (8 GB)Orin Nano~237Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~171Simulate output on iPhone 1720 GB
fits · 19.7 GB free · up to 2K ctx
1 GPU
RX 7900 XT~1714Simulate output on RX 7900 XT32–40 GB
fits · 31.7 GB free · up to 2K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~4160Simulate output on RTX 5090M5 Max36 GB~1150Simulate output on Mac Studio M5 Max (36 GB)M532 GB~384Simulate output on MacBook Air 13" M5 (32 GB)Orin~475Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~313Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at FP16
Similar models
1| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| SmolLM2 135MHugging Face | — | 0.6 GB | ~1524 | same size |
Specifications
- Publisher
- optimum-intel-internal-testing
- Parameters
- 125M
- Architecture
- OPTForCausalLM
- Context Length
- 2,048 tokens
- Vocabulary Size
- 50,272
- Release Date
- 2025-10-21
- License
- Other
Frequently Asked Questions
How much VRAM does Opt 125M need?
Opt 125M requires 0.3 GB of VRAM at FP16.
VRAM = Weights + KV Cache + Overhead
Weights = 125M × 16 bits ÷ 8 = 0.3 GB
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
VRAM usage by quantization
Can I run Opt 125M on a Mac?
Yes — Mac Mini M1 (8 GB) and 155 other Macs can run Opt 125M. Apple Silicon uses unified memory, so the model shares RAM with the system. At FP16 you need at least 0.3 GB of usable unified memory (RAM minus macOS overhead).
Can I run Opt 125M locally?
Yes — Opt 125M can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Opt 125M?
At FP16, Opt 125M can reach ~17143 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~2340 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 0.28 × 0.65 = ~18571 tok/s
Estimated speed at FP16 (0.3 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Opt 125M?
At FP16, the download is about 0.25 GB.
Which GPUs can run Opt 125M?
52 consumer GPUs can run Opt 125M at FP16. Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 52 GPUs have plenty of headroom for comfortable inference.
Which devices can run Opt 125M?
176 devices with unified memory can run Opt 125M at FP16 (0.3 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Apple iPhone 17 Pro, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.