Starcoder2 15B — Hardware Requirements & GPU Compatibility
ChatCode
Starcoder2 15B is a 16.0B-parameter open language model from BigCode in the StarCoder family. It supports a context window of up to 16,384 tokens. At BF16 it needs about 32.38 GB of VRAM — see which GPUs and Macs can run it below.
8.4K downloads
671 likesSpecifications
- Publisher
- BigCode
- Family
- StarCoder
- Parameters
- 16.0B
- Architecture
- Starcoder2ForCausalLM
- Context Length
- 16,384 tokens
- Vocabulary Size
- 49,152
- Release Date
- 2024-06-05
- License
- bigcode-openrail-m
Get Started
HuggingFace
How Much VRAM Does Starcoder2 15B Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| BF16 | 16.00 | 32.4 GB | 33.6 GB | 31.92 GB | Brain floating point 16 — preferred for training |
Which GPUs Can Run Starcoder2 15B?
BF16 · 32.4 GBStarcoder2 15B (BF16) requires 32.4 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 43+ GB is recommended. Using the full 16K context window can add up to 1.2 GB, bringing total usage to 33.6 GB. No single GPU has enough memory — multi-GPU or cluster setups are needed.
Which Devices Can Run Starcoder2 15B?
BF16 · 32.4 GB13 devices with unified memory can run Starcoder2 15B, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Studio M4 Max (36 GB).
Runs great
— Plenty of headroom
Benchmarks
View all 2 →Related Models
Derivatives (1)
Frequently Asked Questions
- How much VRAM does Starcoder2 15B need?
Starcoder2 15B requires 32.4 GB of VRAM at BF16. Full 16K context adds up to 1.2 GB (33.6 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 16.0B × 16 bits ÷ 8 = 31.9 GB
KV Cache + Overhead ≈ 0.5 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 1.7 GB (at full 16K context)
VRAM usage by quantization
BF1632.4 GBBF16 + full context33.6 GB- Can NVIDIA GeForce RTX 5090 run Starcoder2 15B?
No — Starcoder2 15B requires at least 32.4 GB at BF16, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
- Can I run Starcoder2 15B on a Mac?
Starcoder2 15B requires at least 32.4 GB at BF16, 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 Starcoder2 15B locally?
Yes — Starcoder2 15B can run locally on consumer hardware. At BF16 quantization it needs 32.4 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Starcoder2 15B?
At BF16, Starcoder2 15B can reach ~90 tok/s on AMD Instinct MI300X. 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 ÷ 32.4 × 0.55 = ~90 tok/s
Estimated speed at BF16 (32.4 GB)
~90 tok/s~67 tok/s~56 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of Starcoder2 15B?
At BF16, the download is about 31.92 GB.
- Which GPUs can run Starcoder2 15B?
No single consumer GPU has enough VRAM to run Starcoder2 15B at BF16 (32.4 GB). Multi-GPU or professional hardware is required.
- Which devices can run Starcoder2 15B?
13 devices with unified memory can run Starcoder2 15B at BF16 (32.4 GB), including Mac Mini M4 Pro (48 GB), Mac Pro M2 Ultra (192 GB), Mac Studio M2 Ultra (192 GB), Mac Studio M4 Max (128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.