MiniLLM Gpt2 340M — Hardware Requirements & GPU Compatibility
Chat
MiniLLM Gpt2 340M is a 340M-parameter open language model from MiniLLM. It supports a context window of up to 1,024 tokens. At Q4_K_M it needs about 0.22 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M)Best on 24 GB (Q4_K_M · ~2978 tok/s)Macs from 8 GB (Q4_K_M · ~326 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run MiniLLM Gpt2 340M
One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.
Recommended
vllm serve MiniLLM/MiniLLM-gpt2-340M --max-model-len 1024vLLM repo with its own quantisation
Download: Weights on Hugging Face
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX 3080 Ti (12 GB) (11 GB headroom) · Q4_K_M
- Generation speed
- ~2696 tok/s
- generation speed
- Cost per 1M output tokens
- $0.013
- per 1M output tokens
How Much VRAM Does MiniLLM Gpt2 340M Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 0.2 GB | — | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: — context · ~4095 tok/s | |
| 0.2 GB | — | 3-bit medium — 3-bit medium quantization | Fits: — context · ~3640 tok/s | |
| 0.2 GB | — | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: — context · ~2978 tok/s | |
| 0.3 GB | — | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: — context · ~2427 tok/s | |
| 0.3 GB | — | 6-bit, very good — 6-bit quantization, very good quality | Fits: — context · ~2114 tok/s | |
| 0.4 GB | — | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: — context · ~1771 tok/s | |
| 0.8 GB | — | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: — context · ~874 tok/s |
Selected Q4_K_M
4-bit medium quantization0.2GBat 1K context
- Weights
- 0.1 GB
- KV cache
- 0.0 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 23.8 GB free · ~2978 tok/s
≥50 tok/s needs ≥17 GB/s · RTX 5090 ~5295, RTX 3090 Ti ~2978
Which Hardware Can Run MiniLLM Gpt2 340M?
Q4_K_M · 0.2 GBMiniLLM Gpt2 340M (Q4_K_M) requires 0.2 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. 163 devices with unified memory can run MiniLLM Gpt2 340M, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 7.8 GB free · up to — ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~2246Simulate output on RTX 3080RTX 2080 Ti~1820Simulate output on RTX 2080 TiRTX 3070 Ti~1797Simulate output on RTX 3070 TiM3*8 GB~326Simulate output on iMac 24" M3 (8 GB)Orin Nano~301Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~217Simulate output on iPhone 1720 GB
fits · 19.8 GB free · up to — ctx
1 GPU
RX 7900 XT~2182Simulate output on RX 7900 XT32–40 GB
fits · 31.8 GB free · up to — ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~5295Simulate output on RTX 5090M5 Max36 GB~1464Simulate output on Mac Studio M5 Max (36 GB)M532 GB~489Simulate output on MacBook Air 13" M5 (32 GB)Orin~605Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~399Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Granite 4.0 350MIBM | — | 1 GB | ~1150 | same size |
| SmolLM 360MHugging Face | — | 0.6 GB | ~1092 | same size |
| MiniCPM5 2B DSparkOpenBMB | — | 0.6 GB | ~1310 | same size |
| LFM2.5 350MLiquid AI | — | 0.8 GB | ~1130 | same size |
| ERNIE 4.5 0.3B PTBaidu | — | 0.7 GB | ~1213 | same size |
| LFM2 350MLiquid AI | — | 0.8 GB | ~1130 | same size |
Specifications
- Publisher
- MiniLLM
- Parameters
- 340M
- Architecture
- GPT2LMHeadModel
- Context Length
- 1,024 tokens
- Vocabulary Size
- 50,257
- Release Date
- 2024-09-26
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does MiniLLM Gpt2 340M need?
MiniLLM Gpt2 340M requires 0.2 GB of VRAM at Q4_K_M, or 0.8 GB at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 340M × 4.8 bits ÷ 8 = 0.2 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
What's the best quantization for MiniLLM Gpt2 340M?
For MiniLLM Gpt2 340M, Q4_K_M (0.2 GB) offers the best balance of quality and VRAM usage. Q5_K_M (0.3 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 0.2 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run MiniLLM Gpt2 340M on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run MiniLLM Gpt2 340M. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 0.2 GB of usable unified memory (RAM minus macOS overhead).
Can I run MiniLLM Gpt2 340M locally?
Yes — MiniLLM Gpt2 340M can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is MiniLLM Gpt2 340M?
At Q4_K_M, MiniLLM Gpt2 340M can reach ~21818 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~2978 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 ÷ 0.22 × 0.65 = ~23636 tok/s
Estimated speed at Q4_K_M (0.2 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of MiniLLM Gpt2 340M?
At Q4_K_M, the download is about 0.20 GB. The full-precision BF16 version is 0.68 GB. The smallest option (Q2_K) is 0.14 GB.
Which GPUs can run MiniLLM Gpt2 340M?
52 consumer GPUs can run MiniLLM Gpt2 340M at Q4_K_M. 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 MiniLLM Gpt2 340M?
163 devices with unified memory can run MiniLLM Gpt2 340M at Q4_K_M (0.2 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (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.