What are the hardware requirements for Hayabusa 9B?

Hayabusa 9B requires approximately 17.4 GB of VRAM and supports a context window of 256K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Hayabusa 9B and how large is it?

Hayabusa 9B is developed by georvn7, a model with 9b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Hayabusa 9B?

Hayabusa 9B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Hayabusa 9B by georvn7 — VRAM 17.4GB, 256K context

Name: Hayabusa 9B
Author: georvn7

Hayabusa 9B is an open-source language model by georvn7. Features: 9b LLM, VRAM: 17.4GB, Context: 256K, License: apache-2.0.

Agent Base model:finetune:qwen/qwen3... Base model:qwen/qwen3.5-9b Conversational Debugging Dpo En Endpoints compatible Full-finetune Hayabusa Qwen Qwen3.5 Qwen3 5 text Region:us Safetensors Sft Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/georvn7/hayabusa-9b

Hayabusa 9B Parameters and Internals

LLM Name	Hayabusa 9B
Repository 🤗	https://huggingface.co/georvn7/hayabusa-9b
Model Name	/workspace/agent-training/trainer/qwen35_9b_fullft/runs/20260911_154647_contract_hayabusa-9b-temporal-v2-step1-r3-c8-rl_rl/artifacts/full_model
Base Model(s)	Qwen/Qwen3.5-9B Qwen/Qwen3.5-9B
Model Size	9b
Required VRAM	17.4 GB
Updated	2026-09-13
Maintainer	georvn7
Model Type	qwen3_5_text
Model Files	2.0 GB: 1-of-32 2.0 GB: 2-of-32 0.5 GB: 3-of-32 0.5 GB: 4-of-32 0.5 GB: 5-of-32 0.5 GB: 6-of-32 0.4 GB: 7-of-32 0.4 GB: 8-of-32 0.5 GB: 9-of-32 0.4 GB: 10-of-32 0.4 GB: 11-of-32 0.4 GB: 12-of-32 0.5 GB: 13-of-32 0.4 GB: 14-of-32 0.5 GB: 15-of-32 0.5 GB: 16-of-32 0.5 GB: 17-of-32 0.4 GB: 18-of-32 0.4 GB: 19-of-32 0.5 GB: 20-of-32 0.5 GB: 21-of-32 0.4 GB: 22-of-32 0.4 GB: 23-of-32 0.5 GB: 24-of-32 0.4 GB: 25-of-32 0.4 GB: 26-of-32 0.4 GB: 27-of-32 0.5 GB: 28-of-32 0.4 GB: 29-of-32 0.5 GB: 30-of-32 0.5 GB: 31-of-32 0.3 GB: 32-of-32
Supported Languages	en
Model Architecture	Qwen3_5ForCausalLM
License	apache-2.0
Context Length	262144
Model Max Length	262144
Transformers Version	5.3.0.dev0
Tokenizer Class	TokenizersBackend
Padding Token	<\|endoftext\|>
Vocabulary Size	248320
Errors	replace

Rank the Hayabusa 9B capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to Hayabusa 9B

Best Alternatives	Context / RAM	Downloads	Likes
NeoHorse 1 9B	256K / 18 GB	1250	196
Le Gros Chaton	256K / 17.9 GB	235	1
ViAble 1.7 9B	256K / 17.9 GB	1118	3
...isan Translator V1 Qwen 3.5 9B	256K / 17.9 GB	559	0
Estragon 9B	256K / 17.9 GB	421	0
Dot	256K / 17.9 GB	311	1
Merlin Agent	256K / 17.9 GB	571	9
FuseLLM 9B	256K / 17.9 GB	574	0
Carnice 9B	256K / 17.9 GB	358	186
....5 9B Merged V5.3 SFT Selected	256K / 17.9 GB	110	0

Note: green Score (e.g. "73.2") means that the model is better than georvn7/hayabusa-9b.