Smaug Flash — Hardware Requirements & GPU Compatibility
ChatCode
Smaug Flash is a 304.2B-parameter open language model from abacusai. It supports a context window of up to 1,048,576 tokens. At BF16 it needs about 608.68 GB of VRAM — see which GPUs and Macs can run it below.
Smaug Flash runs on a multi-GPU server
BF16 needs ~608.8 GB · about 5× H200 SXM 141 GB
MIT
Not set up for it?
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Smaug Flash
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
vllm serve abacusai/Smaug-Flash --max-model-len 16384 --tensor-parallel-size 8Size the nodes from the table above
Download: Weights on Hugging Face
Other ways
each opens step-by-stepRun in cloud
Rent 5× H200 SXM · BF16
- Estimated hourly price
- ~$17.95/hr
- est., whole server
- GPUs
- 5 GPUs
- 705 GB of VRAM
How Much VRAM Does Smaug Flash Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | Runs on | |
|---|---|---|---|---|
| 608.7 GB | 609 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Runs on 5× H200 SXM 141 GB |
Selected BF16
Brain floating point 16608.8GBat 16K context
- Weights
- 608.4 GB
- KV cache
- 0.2 GB
- Overhead
- 0.3 GB
Server-class: ~608.8 GB at 16K context · about 5× H200 SXM 141 GB
Which Hardware Can Run Smaug Flash?
BF16 · 608.7 GBSmaug Flash (BF16) requires 608.7 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 792+ GB is recommended. Using the full 1049K context window can add up to 11.5 GB, bringing total usage to 620.2 GB. No consumer GPU has enough memory.
Needs server-class hardware
Smaug Flash runs on a multi-GPU server · BF16 needs ~608.8 GB
| Hardware | Quant | Memory | Context | Hardware page | |
|---|---|---|---|---|---|
| 5× H200 SXM 141 GB · 705 GB | BF16 | 608.8 GB | 1M ctx | H200 SXM → | BF16 · 608.8 GB · 1M ctx |
| 4× B200 192 GB · 768 GB | BF16 | 608.8 GB | 1M ctx | B200 → | BF16 · 608.8 GB · 1M ctx |
Also: NVIDIA DGX A100 640GB · NVIDIA DGX H100
By VRAM tier · typical GPUs and Macs at BF16
Consumer GPUs (up to 32 GB) and Macs up to 512 GB: too small for this model (needs ≥ 608.8 GB)8 GB tier →12 GB tier →16 GB tier →24 GB tier →Macs →
How good is Smaug Flash?
llmrun Score blends public benchmark boards into one number; 100 = the average open model.
Not enough data yet for an llmrun Score.
A Score needs results on enough independent boards. How the Composite Score is built
Benchmarks
The 6 boards that weigh most in the llmrun Score, of 10 with a result for this model; badges show who ran the test.
- LiveBench Coding
LiveBench · Jun 2026Independent
Independent77.1
#8/19 open
- LiveBench Reasoning
LiveBench · Jun 2026Independent
Independent86.2
#5/19 open
- LiveBench Math
LiveBench · Jun 2026Independent
Independent90.1
#4/19 open
- LiveBench 2026 Coding
LiveBench · Sep 2026Independent
Independent60.1
#10/15 open
- LiveBench 2026 Long-Horizon
LiveBench · Sep 2026Independent
Independent27.1
#8/15 open
- LiveBench 2026 Design
LiveBench · Sep 2026Independent
Independent76.2
#6/15 open
Similar models
5| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| GLM 5.3 FlashZ.ai | 124.2 | 205.2 GB | —needs > 24 GB | same size |
| MiMo V2.6 Flash MOPDXiaomi | — | 186.5 GB | —needs > 24 GB | same size |
| DeepSeek V4 Flash Vision ExpDeepSeek | — | 183.3 GB | —needs > 24 GB | same size |
| Hy3Tencent | — | 182.3 GB | —needs > 24 GB | same size |
| Step3StepFun | — | 219.1 GB | —needs > 24 GB | same size |
Specifications
- Publisher
- abacusai
- Parameters
- 304.2B
- Active parameters
- 27.4B per token
- Architecture
- DeepseekV4ForCausalLM
- Context Length
- 1,048,576 tokens
- Vocabulary Size
- 129,280
- Release Date
- 2026-08-29
- License
- MIT
Frequently Asked Questions
How much VRAM does Smaug Flash need?
Smaug Flash requires 608.7 GB of VRAM at BF16. Full 1049K context adds up to 11.5 GB (620.2 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 304.2B × 16 bits ÷ 8 = 608.4 GB
KV Cache + Overhead ≈ 0.3 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
KV Cache + Overhead ≈ 11.8 GB (at full 1049K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 5090 run Smaug Flash?
No — Smaug Flash requires at least 608.7 GB at BF16, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
Can I run Smaug Flash on a Mac?
Smaug Flash requires at least 608.7 GB at BF16. None of the 146 Macs we list has that much usable memory (the largest has 508.5 GB after macOS overhead).
Can I run Smaug Flash locally?
Yes — Smaug Flash can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
What's the download size of Smaug Flash?
At BF16, the download is about 608.36 GB.
Which GPUs can run Smaug Flash?
No single consumer GPU has enough VRAM to run Smaug Flash at BF16 (608.7 GB). Multi-GPU or professional hardware is required.
Which devices can run Smaug Flash?
Smaug Flash requires at least 608.7 GB at BF16, which exceeds the usable memory of every unified-memory device we list (the largest has 508.5 GB).