Tiny LLM — Hardware Requirements & GPU Compatibility
Chat
Tiny LLM is a 13M-parameter open language model from arnir0. It supports a context window of up to 1,024 tokens. At Q4_K_M it needs about 0.31 GB of VRAM — see which GPUs and Macs can run it below.
53.5K downloads
52 likesSpecifications
- Publisher
- arnir0
- Parameters
- 13M
- Architecture
- LlamaForCausalLM
- Context Length
- 1,024 tokens
- Vocabulary Size
- 32,000
- Release Date
- 2024-11-03
- License
- MIT
Get Started
HuggingFace
How Much VRAM Does Tiny LLM Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_K | 3.40 | 0.3 GB | — | 0.01 GB | 2-bit quantization with K-quant improvements |
| Q3_K_S | 3.50 | 0.3 GB | — | 0.01 GB | 3-bit small quantization |
| Q3_K_M | 3.90 | 0.3 GB | — | 0.01 GB | 3-bit medium quantization |
| Q4_K_M | 4.80 | 0.3 GB | — | 0.01 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_M | 5.70 | 0.3 GB | — | 0.01 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_K | 6.60 | 0.3 GB | — | 0.01 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 0.3 GB | — | 0.01 GB | 8-bit quantization, near-lossless |
Which GPUs Can Run Tiny LLM?
Q4_K_M · 0.3 GBTiny LLM (Q4_K_M) 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. 35 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti.
Runs great
— Plenty of headroomWhich Devices Can Run Tiny LLM?
Q4_K_M · 0.3 GB33 devices with unified memory can run Tiny LLM, including NVIDIA DGX H100, NVIDIA DGX A100 640GB.
Runs great
— Plenty of headroomRelated Models
Frequently Asked Questions
- How much VRAM does Tiny LLM need?
Tiny LLM requires 0.3 GB of VRAM at Q4_K_M, or 0.3 GB at Q8_0.
VRAM = Weights + KV Cache + Overhead
Weights = 13M × 4.8 bits ÷ 8 = 0 GB
KV Cache + Overhead ≈ 0.3 GB (at 2K context + ~0.3 GB framework)
VRAM usage by quantization
Q4_K_M0.3 GB- What's the best quantization for Tiny LLM?
For Tiny LLM, Q4_K_M (0.3 GB) offers the best balance of quality and VRAM usage. Q5_K_S (0.3 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 0.3 GB.
VRAM requirement by quantization
Q2_K0.3 GBQ3_K_L0.3 GBQ4_K_S0.3 GBQ4_K_M ★0.3 GBQ5_K_M0.3 GBQ8_00.3 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run Tiny LLM on a Mac?
Tiny LLM requires at least 0.3 GB at Q2_K, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run Tiny LLM locally?
Yes — Tiny LLM can run locally on consumer hardware. At Q4_K_M quantization it needs 0.3 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Tiny LLM?
At Q4_K_M, Tiny LLM can reach ~9403 tok/s on AMD Instinct MI300X. On NVIDIA GeForce RTX 4090: ~2114 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: AMD Instinct MI300X → 5300 ÷ 0.3 × 0.55 = ~9403 tok/s
Estimated speed at Q4_K_M (0.3 GB)
~9403 tok/s~2114 tok/s~7028 tok/s~5814 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of Tiny LLM?
At Q4_K_M, the download is about 0.01 GB. The full-precision Q8_0 version is 0.01 GB. The smallest option (Q2_K) is 0.01 GB.
- Which GPUs can run Tiny LLM?
35 consumer GPUs can run Tiny LLM at Q4_K_M (0.3 GB). Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 35 GPUs have plenty of headroom for comfortable inference.
- Which devices can run Tiny LLM?
33 devices with unified memory can run Tiny LLM at Q4_K_M (0.3 GB), including Mac Mini M4 (16 GB), Mac Mini M4 (32 GB), Mac Mini M4 Pro (24 GB), Mac Mini M4 Pro (48 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.