Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

QLoRA fine-tuning 4-bit Gemma 2B on iPhone 15 Pro with MLX Swift. A nice size for fine-tuning on device, getting 70-100 toks/sec depending on the batch. Guide here:

Awni Hannun

34,884 subscribers

19,530 views • 2 years ago •via X (Twitter)

Science & Technology News & Politics Education

Anya Rossi• Live Now

Private livecam show

3 Comments

saurabh 🐳2 years ago

lemme try … this is cool !

Michael Grüner2 years ago

why not

Stijn2 years ago

Are there any plans on integrating (Q)DoRA in the future?

Related Videos

MLX Swift example can also QLoRA fine-tune Llama 3.2. Here's the 1B fine-tuning on my iPhone 15 Pro at > 150 toks/sec. A this rate only takes a few minutes to learn some decent adapters fully on-device.

MLX Swift example can also QLoRA fine-tune Llama 3.2. Here's the 1B fine-tuning on my iPhone 15 Pro at > 150 toks/sec. A this rate only takes a few minutes to learn some decent adapters fully on-device.

Awni Hannun

91,632 views • 1 year ago

Next level: QLoRA fine-tuning 4-bit Llama 3 8B on iPhone 15 pro. Incoming (Q)LoRA MLX Swift example by David Koski: works with lot's of models (Mistral, Gemma, Phi-2, etc)

Next level: QLoRA fine-tuning 4-bit Llama 3 8B on iPhone 15 pro. Incoming (Q)LoRA MLX Swift example by David Koski: works with lot's of models (Mistral, Gemma, Phi-2, etc)

Awni Hannun

581,723 views • 2 years ago

The new Qwen 3.5 by Qwen running on-device on iPhone 17 Pro. Qwen 3.5 beats models 4 times its size, has strong visual understanding, and can toggle reasoning on or off. The 2B 6-bit model here is running with MLX optimized for Apple Silicon.

The new Qwen 3.5 by Qwen running on-device on iPhone 17 Pro. Qwen 3.5 beats models 4 times its size, has strong visual understanding, and can toggle reasoning on or off. The 2B 6-bit model here is running with MLX optimized for Apple Silicon.

Adrien Grondin

3,554,326 views • 4 months ago

Quantized Gemma 2B runs pretty fast on my iPhone 15 pro in MLX Swift. code & docs: Comparable to GPT 3.5 turbo and Mixtral 8x7B in LMSYS Org benchmarks but runs efficiently on an iPhone. Pretty wild.

Quantized Gemma 2B runs pretty fast on my iPhone 15 pro in MLX Swift. code & docs: Comparable to GPT 3.5 turbo and Mixtral 8x7B in LMSYS Org benchmarks but runs efficiently on an iPhone. Pretty wild.

Awni Hannun

79,702 views • 1 year ago

Fine-tuning Mistral 7B with LoRA on a 32 GB M1 (laptop!) in MLX Updated example uses less RAM + support for custom datasets 🚀

Fine-tuning Mistral 7B with LoRA on a 32 GB M1 (laptop!) in MLX Updated example uses less RAM + support for custom datasets 🚀

Awni Hannun

148,141 views • 2 years ago

Alibaba’s Qwen 3.5 is now running fully on-device on the iPhone 17 Pro. It outperforms models 4x its size, delivers strong visual understanding, and lets you switch reasoning on or off. This demo uses the 2B 6-bit version, optimized with MLX for Apple Silicon.

Alibaba’s Qwen 3.5 is now running fully on-device on the iPhone 17 Pro. It outperforms models 4x its size, delivers strong visual understanding, and lets you switch reasoning on or off. This demo uses the 2B 6-bit version, optimized with MLX for Apple Silicon.

Hugging Models

202,723 views • 3 months ago

For even higher throughput and lower latency: batch generation + tensor parallel with mlx-lm + and mlx.distributed. Here it's generating at 63 tok/sec (throughput) with GLM 4.7 in 6-bit and batch size 4 on 4 M3 Ultras:

For even higher throughput and lower latency: batch generation + tensor parallel with mlx-lm + and mlx.distributed. Here it's generating at 63 tok/sec (throughput) with GLM 4.7 in 6-bit and batch size 4 on 4 M3 Ultras:

Awni Hannun

22,721 views • 6 months ago

Fine tuning on the bumps at Sebring 🫨

Fine tuning on the bumps at Sebring 🫨

NTT INDYCAR SERIES

12,580 views • 4 months ago

The new Deep Seek V3 0324 in 4-bit runs at > 20 toks/sec on a 512GB M3 Ultra with mlx-lm!

The new Deep Seek V3 0324 in 4-bit runs at > 20 toks/sec on a 512GB M3 Ultra with mlx-lm!

Awni Hannun

168,870 views • 1 year ago

Ex-NVIDIA engineer who built Unsloth explained RL, kernels, reasoning, quantization, and agents in 2 hours 42 minutes - better than $5000 fine-tuning bootcamps. pick the base model -> write triton kernels for 2x faster fine-tune -> quantize to 4-bit -> run GRPO/DPO -> ship a reasoning model on your single GPU. That loop is why Unsloth is the default way to fine-tune Llama, Qwen, Gemma, and Phi on hardware you already own. Unsloth + Triton kernels + 4-bit quantization + GRPO/DPO + single-GPU fine-tuning - that's the stack. Watch and save it, then fine-tune your first model tonight.

Ex-NVIDIA engineer who built Unsloth explained RL, kernels, reasoning, quantization, and agents in 2 hours 42 minutes - better than $5000 fine-tuning bootcamps. pick the base model -> write triton kernels for 2x faster fine-tune -> quantize to 4-bit -> run GRPO/DPO -> ship a reasoning model on your single GPU. That loop is why Unsloth is the default way to fine-tune Llama, Qwen, Gemma, and Phi on hardware you already own. Unsloth + Triton kernels + 4-bit quantization + GRPO/DPO + single-GPU fine-tuning - that's the stack. Watch and save it, then fine-tune your first model tonight.

h100envy

270,592 views • 1 day ago

intensity, precision, conditioning and passion.. fine tuning the fine tuning 🤼‍♀️

intensity, precision, conditioning and passion.. fine tuning the fine tuning 🤼‍♀️

Nattie

55,627 views • 1 year ago

On-device realtime transcription on iPhone 15 Pro max 🚀 Using MLX-Audio-Swift + Qwen3-ASR-0.6B by Qwen It’s much faster and more consistent with the latest adjustments. Almost ready to push to GH.

On-device realtime transcription on iPhone 15 Pro max 🚀 Using MLX-Audio-Swift + Qwen3-ASR-0.6B by Qwen It’s much faster and more consistent with the latest adjustments. Almost ready to push to GH.

Prince Canuma

44,103 views • 5 months ago

DeepSeek R1 (the 680B MOE) is ~20% faster in the latest mlx / mlx-lm. 4-bit model on 3 M2 Ultras generates 4k tokens at a respectable 15 toks/sec. Plus some QoL improvements: - Only downloads the local shard (much faster startup) - Distributed launcher ships with MLX

DeepSeek R1 (the 680B MOE) is ~20% faster in the latest mlx / mlx-lm. 4-bit model on 3 M2 Ultras generates 4k tokens at a respectable 15 toks/sec. Plus some QoL improvements: - Only downloads the local shard (much faster startup) - Distributed launcher ships with MLX

Awni Hannun

86,921 views • 1 year ago

IN: video fine-tuning support for AI at Meta's V-JEPA 2 in HF transformers 🔥 it comes with > fine-tuning notebook > four models fine-tuned on Diving48 and SSv2 dataset > FastRTC demo on V-JEPA2 SSv2 (see below) we're looking forward to see fine-tuned V-JEPA2 models on Hub ⏯️

IN: video fine-tuning support for AI at Meta's V-JEPA 2 in HF transformers 🔥 it comes with > fine-tuning notebook > four models fine-tuned on Diving48 and SSv2 dataset > FastRTC demo on V-JEPA2 SSv2 (see below) we're looking forward to see fine-tuned V-JEPA2 models on Hub ⏯️

merve

15,625 views • 1 year ago

Llama 3.3 70B 4-bit running on a 128GB M4 Max with MLX LM (~12 toks/sec) with mactop on the right. "Good enough", local, private and open models are now more accessible than ever!

Llama 3.3 70B 4-bit running on a 128GB M4 Max with MLX LM (~12 toks/sec) with mactop on the right. "Good enough", local, private and open models are now more accessible than ever!

Ivan Fioravanti ᯅ

43,929 views • 1 year ago

Fine Tuning with Adrian Martinez

Fine Tuning with Adrian Martinez

John Teigland

21,647 views • 1 year ago

QVAC first demo (android mobile) of its local on-device fine-tuning generalized framework, which aims to bring LORA fine-tuning to any device, scaling from the smallest embedded device to any smartphone, laptop, server, operating system. In combination with WIP multi-node features, we expect QVAC to become the standard for local on-device single/multi node inference and fine-tuning. Imagine a horde of AI agents being able to learn from your emails, notes and messages, maintaining 100℅ privacy, leveraging only your own devices. A P2P private intelligence network at your fingertips. That's how QVAC personal assistant will look like. Future of Intelligence won't be centralized. Infinite Ingelligence vibes.

QVAC first demo (android mobile) of its local on-device fine-tuning generalized framework, which aims to bring LORA fine-tuning to any device, scaling from the smallest embedded device to any smartphone, laptop, server, operating system. In combination with WIP multi-node features, we expect QVAC to become the standard for local on-device single/multi node inference and fine-tuning. Imagine a horde of AI agents being able to learn from your emails, notes and messages, maintaining 100℅ privacy, leveraging only your own devices. A P2P private intelligence network at your fingertips. That's how QVAC personal assistant will look like. Future of Intelligence won't be centralized. Infinite Ingelligence vibes.

Paolo Ardoino 🤖

142,533 views • 8 months ago

Fine-tuning the details.

Fine-tuning the details.

Arsenal Women

195,677 views • 7 months ago

Qwen3 235B MoE (22B active) runs so fast on an M2 Ultra with mlx-lm. - 4-bit model uses ~132GB - Generated 580 tokens at ~28 toks/sec

Qwen3 235B MoE (22B active) runs so fast on an M2 Ultra with mlx-lm. - 4-bit model uses ~132GB - Generated 580 tokens at ~28 toks/sec

Awni Hannun

117,763 views • 1 year ago

Fine-tuning the reflexes 🧤

Fine-tuning the reflexes 🧤

Arsenal

797,597 views • 1 month ago