# Alpamayo-1.5-10B-DFlash by Z Lab: Open-Weight Model Source: https://savrn.com/models/alpamayo-1-5-10b-dflash Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## Runs On What it takes to serve Alpamayo-1.5-10B-DFlash (470M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them. | Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits | | --- | --- | --- | --- | --- | --- | | 16-bit | 0.9 GB | 1.1 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 8-bit | 0.5 GB | 0.6 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 4-bit | 0.2 GB | 0.3 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the [SAVRN Index](https://savrn.com/ai-index/pricing/gpus), read Oct 24, 2026. [Alpamayo-1.5-10B-DFlash on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/alpamayo-1-5-10b-dflash/gpus) ## Model Card Flash Vision-Language-Action Inference for Autonomous Driving DFlash draft model for z-lab/Alpamayo-1.5-10B, used by FlashDrive to accelerate the chain-of-causation reasoning of Alpamayo 1.5. DFlash (ICML 2026) uses a lightweight block-diffusion draft to propose several tokens in parallel; the target verifies each block in a single forward, preserving its output distribution. This draft is a 2-layer Qwen3-style network (block size 8) conditioned on target hidden states from layers 24/30/31/32/34. The repository also ships maskembedding.pt, the trained mask-token embedding FlashDrive appends to the target's embedding table. See the base model card and the FlashDrive repository for the full… Excerpt from the card by Z Lab, licensed other. ## Configuration Architecture DFlashDraftModel Context length (tokens) 40,960 Layers 2 Hidden size 4,096 Feed-forward size 12,288 Attention heads 32 Key/value heads 8 Head dimension 128 Vocabulary size 155,697 RoPE base 1e+06 Model type qwen3 ## Identity and Version Repository z-lab/Alpamayo-1.5-10B-DFlash Publisher Z Lab Task Robotics Modality Control Library Not stated by the source Parameters 470M parameters Languages en Revision 2496c4567acfc043028963daf2d5f73a8c7fc7bc First published 2026-03-31 Last updated 2026-07-04 ## Files and Weights 5 files, 939.6 MB in total. The weights are 2 files totalling 939.6 MB in pt, safetensors. Weights2 files · 939.6 MB Configuration1 file · 926 B Documentation1 file · 3.2 KB Repository1 file · 1.5 KB Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | mask_embedding.pt | Weights | 9.8 KB | 2489e4a6a320 | | model.safetensors | Weights | 939.6 MB | 8616581ab218 | | config.json | Configuration | 926 B | — | | README.md | Documentation | 3.2 KB | — | | .gitattributes | Repository | 1.5 KB | — | ## License and Download License other Access Open weights, no gate Download size 939.6 MB [Download from Z Lab](https://huggingface.co/z-lab/Alpamayo-1.5-10B-DFlash) Released by Z Lab through its official repository on Hugging Face. [Read the license](https://huggingface.co/nvidia/Alpamayo-1.5-10B/blob/main/LICENSE). ## Built From - Derived from [z-lab/Alpamayo-1.5-10B](https://savrn.com/models/alpamayo-1-5-10b-2) - Described by [arXiv:2602.06036](https://savrn.com/papers/dflash-block-diffusion-for-flash-speculative-decoding) ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 939.6 MB | | 16-bit | 0.9 GB | | 8-bit | 0.5 GB | | 4-bit | 0.2 GB | Weights only, from the published parameter count; the key-value cache and runtime add to this. ## Questions About Alpamayo-1.5-10B-DFlash ### How much GPU memory does Alpamayo-1.5-10B-DFlash need? About 1.1 GB at 16-bit and 0.3 GB at 4-bit: the weights (470M parameters) plus a working margin. A long context needs more. ### What is the cheapest GPU to run Alpamayo-1.5-10B-DFlash on? At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists. ### What license is Alpamayo-1.5-10B-DFlash released under? other, as its publisher declares it. Read the license text before commercial use. ### What is Alpamayo-1.5-10B-DFlash's context length? 40,960 tokens, from the maximum position embeddings in its published configuration. ## Similar Models Model · Robotics ### [smolvla_base](https://savrn.com/models/smolvla-base) [LeRobot](https://savrn.com/model-publishers/lerobot) SmolVLA is a compact, efficient Vision-Language-Action (VLA) model designed for affordable robotics, trainable on a single GPU and deployable on consumer hardware, while matching the performance of much larger VLAs through community-driven data. Original paper: (SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics)[https://arxiv.org/abs/2506.01844] For full installation details (including optional video dependencies such as ffmpeg for torchcodec), see the official documentation: https://huggingface.co/docs/lerobot/installation If you’re training / fine-tuning, you typically call forward(...) to get a loss and then: - -policy.chunksize=... - -policy.nactionsteps=...… Open weights apache-2.0 450M parameters lerobot [View model](https://savrn.com/models/smolvla-base) Model · Robotics ### [smolvla_libero](https://savrn.com/models/smolvla-libero-2) [LeRobot](https://savrn.com/model-publishers/lerobot) SmolVLA is a compact, efficient vision-language-action model that achieves competitive performance at reduced computational costs and can be deployed on consumer-grade hardware. This policy has been trained and pushed to the Hub using LeRobot. See the full documentation at LeRobot Docs. For a complete walkthrough, see the training guide. Below is the short version on how to train and run inference/eval: Writes checkpoints to outputs/train/ /checkpoints/. Prefix the dataset repo with eval\ and supply --policy.path pointing to a local or hub checkpoint. Open weights apache-2.0 450M parameters lerobot [View model](https://savrn.com/models/smolvla-libero-2) Model · Robotics ### [smolvla_robotwin](https://savrn.com/models/smolvla-robotwin) [LeRobot](https://savrn.com/model-publishers/lerobot) SmolVLA is a compact, efficient vision-language-action model that achieves competitive performance at reduced computational costs and can be deployed on consumer-grade hardware. This policy has been trained and pushed to the Hub using LeRobot. See the full documentation at LeRobot Docs. For a complete walkthrough, see the training guide. Below is the short version on how to train and run inference/eval: Writes checkpoints to outputs/train/ /checkpoints/. Prefix the dataset repo with eval\ and supply --policy.path pointing to a local or hub checkpoint. Open weights apache-2.0 450M parameters lerobot [View model](https://savrn.com/models/smolvla-robotwin) Model · Robotics ### [smolvla-policy-test](https://savrn.com/models/smolvla-policy-test) [Hayato Nakamura](https://savrn.com/model-publishers/tron-hayato) SmolVLA is a compact, efficient vision-language-action model that achieves competitive performance at reduced computational costs and can be deployed on consumer-grade hardware. This policy has been trained and pushed to the Hub using LeRobot. Learn how to train and run it in the LeRobot smolvla guide, or browse the full documentation. The policy consumes these observation features and produces these action features. Inputs Outputs New to LeRobot? These guides cover the full workflow: - Install LeRobot — set up the lerobot package. - Hardware setup — assemble, wire, and calibrate your robot and cameras. - Record data & train a policy — the end-to-end imitation-learning walkthrough. - CLI… Open weights apache-2.0 450M parameters lerobot [View model](https://savrn.com/models/smolvla-policy-test) Model · Robotics ### [smolvla-policy](https://savrn.com/models/smolvla-policy) [Hayato Nakamura](https://savrn.com/model-publishers/tron-hayato) SmolVLA is a compact, efficient vision-language-action model that achieves competitive performance at reduced computational costs and can be deployed on consumer-grade hardware. This policy has been trained and pushed to the Hub using LeRobot. Learn how to train and run it in the LeRobot smolvla guide, or browse the full documentation. The policy consumes these observation features and produces these action features. Inputs Outputs New to LeRobot? These guides cover the full workflow: - Install LeRobot — set up the lerobot package. - Hardware setup — assemble, wire, and calibrate your robot and cameras. - Record data & train a policy — the end-to-end imitation-learning walkthrough. - CLI… Open weights apache-2.0 450M parameters lerobot [View model](https://savrn.com/models/smolvla-policy) Model · Robotics ### [thali_smolvla](https://savrn.com/models/thali-smolvla) [Prashant Thakur](https://savrn.com/model-publishers/prashant-77) lerobot/smolvlabase fine-tuned on Prashant-77/thaliall (1050 scripted-expert episodes, 7 skills, language-conditioned, 3 cameras). Camera keys are renamed at train and inference time: overhead → camera1, wrista → camera2, wristb → camera3 (--renamemap; runtime/executors.py applies the same map). Checkpoints in this repo. Root = 20 000 total steps. step14000/ = the best per-skill checkpoint (14 000 steps at batch 16 on a T4; the last 6 000 steps ran at batch 4 with a fresh optimizer on a smaller GPU and lost ground). Policy-only success per skill from task-consistent start states, 20 held-out seeds (eval/skilleval.py --kind smolvla): The scripted expert reaches 9/10 on the full task; the… Open weights apache-2.0 450M parameters lerobot [View model](https://savrn.com/models/thali-smolvla) ## Z Lab [All models and datasets](https://savrn.com/model-publishers/z-lab) ## Versions - [2496c4567acf](https://savrn.com/models/alpamayo-1-5-10b-dflash/versions/2496c4567acf) · current 2026-09-18 ## Explore More - [All robotics models](https://savrn.com/models/tasks/robotics) - [Model comparisons](https://savrn.com/models/comparisons) - [The model directory](https://savrn.com/models) - [Open model prices by host](https://savrn.com/ai-index/pricing/open-models) ## Source - Repository metadata, read 2026-10-21. - [Hugging Face record](https://huggingface.co/z-lab/Alpamayo-1.5-10B-DFlash) - [How the hub is built](https://savrn.com/model-hub/methodology)