Anthropic
Claude 5 Fable (480B)
Anthropic's Mythos-class flagship for long-horizon agentic coding
Model Summary
Family
Claude 5
Version
5.0
Parameters
480B (est.)
Parameter counts for closed models are estimates; vendors rarely publish exact sizes.
VRAM Requirements by Quantization
Memory needed to serve Claude 5 Fable (480B) for inference, including a 20% overhead for activations and KV cache.
| Precision | VRAM needed | Smallest single GPU that fits |
|---|---|---|
| INT4 (4-bit) | 268.22 GB | NVIDIA B300 SXM (288 GB) |
| INT8 (8-bit) | 536.44 GB | Multi-GPU required |
| FP16 (16-bit) | 1072.88 GB | Multi-GPU required |
| FP32 (32-bit) | 2145.77 GB | Multi-GPU required |
Recommended GPU Configurations
Cheapest on-demand configurations to serve Claude 5 Fable (480B) at 8-bit (536 GB VRAM).
5x AMD Instinct MI250X
640 GB total VRAM · CDNA 2
~$12.50/h
34x NVIDIA T4
544 GB total VRAM · Turing · multi-node
~$17.00/h
3x AMD Instinct MI300X
576 GB total VRAM · CDNA 3
~$18.00/h
Quick GPU Planning
Use the calculator pre-filled with this exact version to estimate memory, speed, and compute requirements in a few clicks.
Access Pre-filled CalculatorFrequently Asked Questions
How much VRAM do you need to run Claude 5 Fable (480B)?
With an estimated 480B parameters, Claude 5 Fable (480B) needs roughly 536 GB of VRAM in 8-bit (INT8), 268 GB in 4-bit, and 1073 GB in FP16, including a 20% overhead for activations and KV cache.
Which GPUs can run Claude 5 Fable (480B)?
At 8-bit quantization, the most cost-effective option is 5x AMD Instinct MI250X (640 GB combined VRAM, around $12.50/hour on-demand). Higher-end cards like the NVIDIA B200 or AMD MI355X reduce the GPU count needed.
Can Claude 5 Fable (480B) run on a single GPU?
Only with aggressive quantization: in 4-bit, a single NVIDIA B300 SXM (288 GB) can fit it. At 8-bit or higher, you need a multi-GPU setup.
How much does it cost to serve Claude 5 Fable (480B) in the cloud?
Renting 5x Instinct MI250X costs on the order of $12.50/hour, i.e. about $9,125/month running 24/7. Actual prices vary by provider and commitment; spot and reserved capacity can be significantly cheaper.
Deploy on a GPU cloud
Rent 5x Instinct MI250X by the hour instead of buying hardware.
