—per task?
self-host onlyprice
—output speed?
262Kcontext
Alternatives?
ModelScoreCostPer taskWhat changes
bOpenmammoth-1 (27B)+3%—$0 / —Context: 262K → 16K+2ModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
bmammoth-1 (27B)Open+3%—$0 / —Context: 262K → 16K+2ModelScoreCostPer taskWhat changes
vD3 TurboOpensame——Context: 262K → 33KvD3 FlashOpen-19%——vD3 MiniOpen-26%——vvLLM-SR Decision 2.0 Lux 9BOpen-32%——Context: 262K → 16K+2vvLLM-SR Decision 2.0 Nox 4BOpen-39%——Context: 262K → 16K+2vvLLM-SR d3-nano (2B)Open-49%——vvLLM-SR d3-lite (0.8B)Open-61%——vvLLM-SR Decision 2.0 Sol 2BOpen-65%——Context: 262K → 16K+2All vLLM Semantic Router models →Run it yourself
estimateFits on✗ 8 GB (phone, laptop)✓ 16 GB · compressed✓ 24 GB GPU · compressed✓ 48 GB GPU · 8-bit✓ 80 GB GPU · full quality
LicenceApache-2.0free for commercial use
Benchmarks
in Decision modelsFAQ
How much does D3 cost?›
There is no per-unit API price: self-host only.
Is D3 free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is D3, and what is it best at?›
#6 of 99 in Decision models.
Is D3 better than D3 Turbo?›
On Community Decision Index 0.4.1, D3 scores 67.2 vs 67.3 — about the same.
What is the context window of D3?›
262K tokens.
When was D3 released? Is there a newer version?›
Released 2026-10-10. It is the current version in its line.
Is D3 open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run D3 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can D3 read images?›
It accepts text and images, video as input.
What is D3?›
D3 is an open-weights language model from vLLM Semantic Router, released Oct 2026.
What are the best alternatives to D3?›
mammoth-1 (27B) (open weights).