Community derivatives from Deepseek V4 and V4.1
Daniel Rosehill
danielrosehill
AI & ML interests
Speech to text (STT), voice workflows, MCP, AI agents, orchestration, automation.
Recent Activity
updated a collection 1 minute ago
Medical liked a model 2 minutes ago
Falconsai/medical_summarization updated a collection about 2 hours ago
Deepseek V4.1Organizations
Models that apply a text-to-text transformation as a second pass on speech transcribed using ASR.
-
zai-org/GLM-ASR-Nano-2512
Automatic Speech Recognition • 2B • Updated • 59.4k • 389 -
nvidia/canary-qwen-2.5b
Automatic Speech Recognition • 3B • Updated • 28.9k • 463 -
microsoft/Phi-4-multimodal-instruct
Automatic Speech Recognition • 6B • Updated • 257k • 1.61k -
facebook/omniASR-LLM-7B
Automatic Speech Recognition • Updated • 38
The two official variants of Voxtral (audio multimodal model) released by Mistral in July 2025
Open sourced audio datasets for STT/ASR. All recordings by me (Daniel Rosehill) unless otherwise accredited.
Useful datasets and models for ASR projects and fine tuning
-
futo-org/acft-whisper-tiny
Automatic Speech Recognition • 57.7M • Updated • 7 • 2 -
futo-org/acft-whisper-small.en
Automatic Speech Recognition • 0.3B • Updated • 9 • 2 -
futo-org/acft-whisper-base.en
Automatic Speech Recognition • 99.1M • Updated • 55 • 2 -
futo-org/acft-whisper-tiny.en
Automatic Speech Recognition • 57.7M • Updated • 12 • 1
Collection of models gathered together for my wife who is an architect (of buildings!)
-
openai/whisper-base
Automatic Speech Recognition • 72.6M • Updated • 1.54M • 289 -
openai/whisper-base.en
Automatic Speech Recognition • 72.6M • Updated • 264k • 45 -
onnx-community/whisper-base_timestamped
Automatic Speech Recognition • Updated • 16.7k • 32 -
Systran/faster-whisper-base
Automatic Speech Recognition • Updated • 1.66M • 36
-
Pharmacology Graph Explorer
💊3Explore predicted drug‑target and disease links
-
Medical Diagnosis
📉71Classify symptoms to diagnose health issues
-
MediAI Medical AI Agent
🚀28AI-Powered Diagnosis & Treatment Assistant
-
Lisdexamfetamine Split Dose Modeller
🚀Model split-dose protocols for lisdexamfetamine/Vyvanse
Spaces and models that may have applications in open source intelligence (OSINT)
AI use-cases and appliations that I found interesting (a repo for myself to explore!)
LLMs optimised for instruction following rather than conversational use - quants and original models
LLMs optimised for running "on-device" (specifically, for this collection, on smartphones with standard inference capabilities)
LLMs (mostly quants) that are small enough to run locally (on my hardware). My go-tos.
-
Qwen/Qwen3-VL-8B-Instruct
Image-Text-to-Text • 9B • Updated • 19.7M • • 1.14k -
zai-org/GLM-4.6
Text Generation • 357B • Updated • 19.3k • • 1.24k -
Qwen/Qwen3-VL-8B-Thinking
Image-Text-to-Text • 9B • Updated • 162k • • 224 -
meta-llama/Llama-3.1-8B-Instruct
Text Generation • 8B • Updated • 5.91M • • 7.77k
-
danielrosehill/Shakespearean-Text-Transformation-Prompts
Viewer • Updated • 1 • 105 -
danielrosehill/Speech-To-Text-System-Prompts-2
Viewer • Updated • 2 • 306 • 1 -
System Prompt Reformatter
📚Reformats system prompts in the 2nd person and other edits
-
BLUF Email Formatter
📧Format emails with clear subject lines and summaries
Projects related to the city of Jerusalem
ISO standard related projects
Datasets and projects related to sustainability esp impact investing
Datasets for generating characters from 2D/3D
Photo galleries of places in Israel for world generation use cases, among others
Experiments in AI and prompt engineering
-
Max Output Tokens Analysis
📊Display max output tokens for models over time
-
LLM Long Output Experiment (Code Generation)
📈1Evaluating max single output length of code gen LLMs
-
Single Shot Brevity Training
📈Using one example to train an LLM for informational brevity
-
Local STT Eval One Sample
😻Single sample eval for WER on various Whisper models
-
modularai/Llama-3.1-8B-Instruct-GGUF
Text Generation • 8B • Updated • 8.41k • 17 -
MaziyarPanahi/WizardLM-2-7B-GGUF
Text Generation • 7B • Updated • 154k • 83 -
MaziyarPanahi/mathstral-7B-v0.1-GGUF
Text Generation • 7B • Updated • 153k • 8 -
MaziyarPanahi/phi-4-GGUF
Text Generation • 15B • Updated • 155k • 10
Datasets in Hebrew
Collection of ASR models for customized STT model training
-
nvidia/parakeet-tdt-0.6b-v2
Automatic Speech Recognition • Updated • 191k • 1.55k -
ibm-granite/granite-speech-3.3-8b
Automatic Speech Recognition • 9B • Updated • 42.8k • 171 -
nvidia/canary-qwen-2.5b
Automatic Speech Recognition • 3B • Updated • 28.9k • 463 -
facebook/omniASR-W2V-1B
Automatic Speech Recognition • Updated • 6
Speech to text models that I've had the best experience with on my current hardware (One Plus Nord 3 )
-
imvladikon/wav2vec2-large-xlsr-53-hebrew
Automatic Speech Recognition • 0.3B • Updated • 19.5k • 8 -
Mizurodp/wav2vec2-large-xls-r-300m-hebrew-colab
Automatic Speech Recognition • Updated • 17 • 1 -
imvladikon/wav2vec2-xls-r-300m-lm-hebrew
Automatic Speech Recognition • 0.3B • Updated • 72 • 4 -
imvladikon/wav2vec2-xls-r-1b-hebrew
Automatic Speech Recognition • 1.0B • Updated • 451 • 2
Quick collection of models I'm evaluating for cogeneration. Minimum inclusion criteria includes tool usage and MCP. Looking for something fast and goo
Open source models with audio understanding. Tracking mostly vendor releases in the audio and text to text subclassification of multimodal.
Whisper fine tunes for use with FUTO keyboard on Android (training: Modal based on Whisper-ACFT skeleton from FUTO)
Whisper fine-tunes for my voice and vocab (tech, Hebrew). About 1 hour of training data so still very much POCs!
-
danielrosehill/daniel_whisper_finetune_large_v3_turbo_v2
Automatic Speech Recognition • 0.8B • Updated • 21 -
danielrosehill/daniel_whisper_finetune_medium_v2
Automatic Speech Recognition • 0.8B • Updated • 18 -
danielrosehill/daniel_whisper_finetune_tiny_v2
Automatic Speech Recognition • 37.8M • Updated • 15 -
danielrosehill/daniel_whisper_finetune_base_v2
Automatic Speech Recognition • 72.6M • Updated • 17
-
nvidia/parakeet-tdt-0.6b-v3
Automatic Speech Recognition • 0.6B • Updated • 577k • • 1.14k -
ibm-granite/granite-speech-3.3-8b
Automatic Speech Recognition • 9B • Updated • 42.8k • 171 -
facebook/seamless-m4t-v2-large
Automatic Speech Recognition • 2B • Updated • 306k • 1.01k -
facebook/wav2vec2-base-960h
Automatic Speech Recognition • 94.4M • Updated • 1.45M • 404
Fine tune of Whisper Large (V3, Turbo) using a small corpus of mixed language English sentences with Hebrew to improve accuracy
Useful models for demonstrating what early and pre-Transformer LLMs looked and functioned like
-
danielrosehill/Podcast-ASR-Evaluation
Viewer • Updated • 27 • 14 -
danielrosehill/Long-Prompt-Experiment
Viewer • Updated • 92 • 82 -
Podcast ASR Evaluation
🎙ASR benchmark comparing local and cloud models
-
LLM Long Output Experiment (Code Generation)
📈1Evaluating max single output length of code gen LLMs
Whisper fine-tuned on 1 hour of my voice
Models that work in unison with core STT models in voice workflows
-
pyannote/voice-activity-detection
Automatic Speech Recognition • Updated • 1.21M • 242 -
pyannote/speaker-diarization-3.1
Automatic Speech Recognition • Updated • 7.95M • 3.78k -
pyannote/overlapped-speech-detection
Automatic Speech Recognition • Updated • 9.42k • 64 -
pipecat-ai/smart-turn-v3
Voice Activity Detection • Updated • 204
-
unsloth/whisper-large-v3
Automatic Speech Recognition • 2B • Updated • 2.95k • 20 -
unsloth/whisper-small
Automatic Speech Recognition • 0.2B • Updated • 790 • 7 -
unsloth/CrisperWhisper
Automatic Speech Recognition • 2B • Updated • 113 • 16 -
unsloth/whisper-large-v3-turbo
Automatic Speech Recognition • 0.8B • Updated • 5.96k • 12
Just for fun projects for converting conventional text into Shakespearean English
Experimentary workflows (mostly) that aim to rethink how contextual data can be gathered to achieve personalised inference in AI systems
-
tencent/HunyuanImage-3.0
Text-to-Image • 83B • Updated • 3.17k • • 1.13k -
black-forest-labs/FLUX.1-schnell
Text-to-Image • 12B • Updated • 574k • • 5.94k -
stabilityai/stable-diffusion-xl-base-1.0
Text-to-Image • 3B • Updated • 3.09M • • 8.2k -
Qwen/Qwen-Image
Text-to-Image • 20B • Updated • 308k • • 2.65k
Quantized versions of image generation models
AI agent network configs and projects
Projects related to Israel
Datasets for STT training, classification fine-tuning - especially voice notes
Collections of system prompts for assistants, agents, and workflows
Version controlled refactors for data analysis of the Global Value Factor Database (GVFD) by the International Foundation for Valuing Impacts (IFVI).
Implementations of apps for simple text reformatting tasks
Spaces to do with Hebrew language AI
Language models supporting TTS (predominantly the text generation task). These are almost exclusively fine-tunes of Gemma, Mistral, or the Qwen models
Community derivatives from Deepseek V4 and V4.1
Models that apply a text-to-text transformation as a second pass on speech transcribed using ASR.
Speech to text models that I've had the best experience with on my current hardware (One Plus Nord 3 )
-
imvladikon/wav2vec2-large-xlsr-53-hebrew
Automatic Speech Recognition • 0.3B • Updated • 19.5k • 8 -
Mizurodp/wav2vec2-large-xls-r-300m-hebrew-colab
Automatic Speech Recognition • Updated • 17 • 1 -
imvladikon/wav2vec2-xls-r-300m-lm-hebrew
Automatic Speech Recognition • 0.3B • Updated • 72 • 4 -
imvladikon/wav2vec2-xls-r-1b-hebrew
Automatic Speech Recognition • 1.0B • Updated • 451 • 2
Quick collection of models I'm evaluating for cogeneration. Minimum inclusion criteria includes tool usage and MCP. Looking for something fast and goo
-
zai-org/GLM-ASR-Nano-2512
Automatic Speech Recognition • 2B • Updated • 59.4k • 389 -
nvidia/canary-qwen-2.5b
Automatic Speech Recognition • 3B • Updated • 28.9k • 463 -
microsoft/Phi-4-multimodal-instruct
Automatic Speech Recognition • 6B • Updated • 257k • 1.61k -
facebook/omniASR-LLM-7B
Automatic Speech Recognition • Updated • 38
The two official variants of Voxtral (audio multimodal model) released by Mistral in July 2025
Open source models with audio understanding. Tracking mostly vendor releases in the audio and text to text subclassification of multimodal.
Whisper fine tunes for use with FUTO keyboard on Android (training: Modal based on Whisper-ACFT skeleton from FUTO)
Open sourced audio datasets for STT/ASR. All recordings by me (Daniel Rosehill) unless otherwise accredited.
Whisper fine-tunes for my voice and vocab (tech, Hebrew). About 1 hour of training data so still very much POCs!
-
danielrosehill/daniel_whisper_finetune_large_v3_turbo_v2
Automatic Speech Recognition • 0.8B • Updated • 21 -
danielrosehill/daniel_whisper_finetune_medium_v2
Automatic Speech Recognition • 0.8B • Updated • 18 -
danielrosehill/daniel_whisper_finetune_tiny_v2
Automatic Speech Recognition • 37.8M • Updated • 15 -
danielrosehill/daniel_whisper_finetune_base_v2
Automatic Speech Recognition • 72.6M • Updated • 17
-
nvidia/parakeet-tdt-0.6b-v3
Automatic Speech Recognition • 0.6B • Updated • 577k • • 1.14k -
ibm-granite/granite-speech-3.3-8b
Automatic Speech Recognition • 9B • Updated • 42.8k • 171 -
facebook/seamless-m4t-v2-large
Automatic Speech Recognition • 2B • Updated • 306k • 1.01k -
facebook/wav2vec2-base-960h
Automatic Speech Recognition • 94.4M • Updated • 1.45M • 404
Useful datasets and models for ASR projects and fine tuning
Fine tune of Whisper Large (V3, Turbo) using a small corpus of mixed language English sentences with Hebrew to improve accuracy
Useful models for demonstrating what early and pre-Transformer LLMs looked and functioned like
-
futo-org/acft-whisper-tiny
Automatic Speech Recognition • 57.7M • Updated • 7 • 2 -
futo-org/acft-whisper-small.en
Automatic Speech Recognition • 0.3B • Updated • 9 • 2 -
futo-org/acft-whisper-base.en
Automatic Speech Recognition • 99.1M • Updated • 55 • 2 -
futo-org/acft-whisper-tiny.en
Automatic Speech Recognition • 57.7M • Updated • 12 • 1
-
danielrosehill/Podcast-ASR-Evaluation
Viewer • Updated • 27 • 14 -
danielrosehill/Long-Prompt-Experiment
Viewer • Updated • 92 • 82 -
Podcast ASR Evaluation
🎙ASR benchmark comparing local and cloud models
-
LLM Long Output Experiment (Code Generation)
📈1Evaluating max single output length of code gen LLMs
Whisper fine-tuned on 1 hour of my voice
Models that work in unison with core STT models in voice workflows
-
pyannote/voice-activity-detection
Automatic Speech Recognition • Updated • 1.21M • 242 -
pyannote/speaker-diarization-3.1
Automatic Speech Recognition • Updated • 7.95M • 3.78k -
pyannote/overlapped-speech-detection
Automatic Speech Recognition • Updated • 9.42k • 64 -
pipecat-ai/smart-turn-v3
Voice Activity Detection • Updated • 204
Collection of models gathered together for my wife who is an architect (of buildings!)
-
unsloth/whisper-large-v3
Automatic Speech Recognition • 2B • Updated • 2.95k • 20 -
unsloth/whisper-small
Automatic Speech Recognition • 0.2B • Updated • 790 • 7 -
unsloth/CrisperWhisper
Automatic Speech Recognition • 2B • Updated • 113 • 16 -
unsloth/whisper-large-v3-turbo
Automatic Speech Recognition • 0.8B • Updated • 5.96k • 12
-
openai/whisper-base
Automatic Speech Recognition • 72.6M • Updated • 1.54M • 289 -
openai/whisper-base.en
Automatic Speech Recognition • 72.6M • Updated • 264k • 45 -
onnx-community/whisper-base_timestamped
Automatic Speech Recognition • Updated • 16.7k • 32 -
Systran/faster-whisper-base
Automatic Speech Recognition • Updated • 1.66M • 36
Just for fun projects for converting conventional text into Shakespearean English
Experimentary workflows (mostly) that aim to rethink how contextual data can be gathered to achieve personalised inference in AI systems
-
Pharmacology Graph Explorer
💊3Explore predicted drug‑target and disease links
-
Medical Diagnosis
📉71Classify symptoms to diagnose health issues
-
MediAI Medical AI Agent
🚀28AI-Powered Diagnosis & Treatment Assistant
-
Lisdexamfetamine Split Dose Modeller
🚀Model split-dose protocols for lisdexamfetamine/Vyvanse
Spaces and models that may have applications in open source intelligence (OSINT)
AI use-cases and appliations that I found interesting (a repo for myself to explore!)
LLMs optimised for instruction following rather than conversational use - quants and original models
-
tencent/HunyuanImage-3.0
Text-to-Image • 83B • Updated • 3.17k • • 1.13k -
black-forest-labs/FLUX.1-schnell
Text-to-Image • 12B • Updated • 574k • • 5.94k -
stabilityai/stable-diffusion-xl-base-1.0
Text-to-Image • 3B • Updated • 3.09M • • 8.2k -
Qwen/Qwen-Image
Text-to-Image • 20B • Updated • 308k • • 2.65k
Quantized versions of image generation models
LLMs optimised for running "on-device" (specifically, for this collection, on smartphones with standard inference capabilities)
LLMs (mostly quants) that are small enough to run locally (on my hardware). My go-tos.
-
Qwen/Qwen3-VL-8B-Instruct
Image-Text-to-Text • 9B • Updated • 19.7M • • 1.14k -
zai-org/GLM-4.6
Text Generation • 357B • Updated • 19.3k • • 1.24k -
Qwen/Qwen3-VL-8B-Thinking
Image-Text-to-Text • 9B • Updated • 162k • • 224 -
meta-llama/Llama-3.1-8B-Instruct
Text Generation • 8B • Updated • 5.91M • • 7.77k
-
danielrosehill/Shakespearean-Text-Transformation-Prompts
Viewer • Updated • 1 • 105 -
danielrosehill/Speech-To-Text-System-Prompts-2
Viewer • Updated • 2 • 306 • 1 -
System Prompt Reformatter
📚Reformats system prompts in the 2nd person and other edits
-
BLUF Email Formatter
📧Format emails with clear subject lines and summaries
AI agent network configs and projects
Projects related to the city of Jerusalem
Projects related to Israel
ISO standard related projects
Datasets and projects related to sustainability esp impact investing
Datasets for STT training, classification fine-tuning - especially voice notes
Datasets for generating characters from 2D/3D
Collections of system prompts for assistants, agents, and workflows
Photo galleries of places in Israel for world generation use cases, among others
Version controlled refactors for data analysis of the Global Value Factor Database (GVFD) by the International Foundation for Valuing Impacts (IFVI).
Experiments in AI and prompt engineering
-
Max Output Tokens Analysis
📊Display max output tokens for models over time
-
LLM Long Output Experiment (Code Generation)
📈1Evaluating max single output length of code gen LLMs
-
Single Shot Brevity Training
📈Using one example to train an LLM for informational brevity
-
Local STT Eval One Sample
😻Single sample eval for WER on various Whisper models
Implementations of apps for simple text reformatting tasks
-
modularai/Llama-3.1-8B-Instruct-GGUF
Text Generation • 8B • Updated • 8.41k • 17 -
MaziyarPanahi/WizardLM-2-7B-GGUF
Text Generation • 7B • Updated • 154k • 83 -
MaziyarPanahi/mathstral-7B-v0.1-GGUF
Text Generation • 7B • Updated • 153k • 8 -
MaziyarPanahi/phi-4-GGUF
Text Generation • 15B • Updated • 155k • 10
Spaces to do with Hebrew language AI
Datasets in Hebrew
Collection of ASR models for customized STT model training
-
nvidia/parakeet-tdt-0.6b-v2
Automatic Speech Recognition • Updated • 191k • 1.55k -
ibm-granite/granite-speech-3.3-8b
Automatic Speech Recognition • 9B • Updated • 42.8k • 171 -
nvidia/canary-qwen-2.5b
Automatic Speech Recognition • 3B • Updated • 28.9k • 463 -
facebook/omniASR-W2V-1B
Automatic Speech Recognition • Updated • 6
Language models supporting TTS (predominantly the text generation task). These are almost exclusively fine-tunes of Gemma, Mistral, or the Qwen models