petergilani

Macaron-V1-Tall-L2-Coding-Fused-BF16

🔍 What Was Fused

Table with columns: Component, Status
Component	Status
Base model	`Macaron-V1-Tall` (Qwen3.6-35B-A3B, BF16) ✅
L2 Coding LoRA (rank 64)	Fused via `W_fused = W_base + (B @ A)` ✅
L0 Chat LoRA	❌ Not included (remains an adapter)
L1 Agent LoRA	❌ Not included (remains an adapter)
L3 GenUI LoRA	❌ Not included (remains an adapter)
MTP module	❌ Removed (`mtp_num_hidden_layers = 0`)

Verification Details

Fusion was confirmed by inspecting model.safetensors.index.json:

No LoRA keys: The weight map contains only standard base model keys (linear_attn.*, self_attn.*, mlp.switch_mlp.*, mlp.shared_expert.*, mlp.gate.*). There are zero lora_A, lora_B, or adapter_* keys — confirming all L2 LoRA weights were absorbed into the base.
Parameter count: total_parameters = 34,660,608,768 (~34.7B) — matches the Macaron-V1-Tall base exactly, since a rank-64 LoRA adds negligible parameters that are now embedded in the base weights.
MTP stripped: mtp_num_hidden_layers = 0 confirms the Multi-Token Prediction module was removed as documented in the fusion plan.
Sharding: 14 safetensors files totaling ~64.5 GB at BF16 precision.

🏗️ Architecture Summary

Table with columns: Field, Value
Field	Value
Base model	Macaron-V1-Tall (Qwen3.6-35B-A3B)
Fusion method	Linear combination: `W_fused = W_base + (B @ A)`
LoRA rank	64
LoRA target	L2 Coding specialist
Context length	262K
Layers	40
Hidden size	2048
Attention heads	16

What the L2 Coding LoRA Adds

The L2 Coding specialist was trained for:

Repository-level coding — understanding project context, navigating codebases
SWE tasks — bug fixing, feature implementation, refactoring
Terminal use — shell commands, git operations, build/test workflows
Code-native workflows — tool calling, code generation, documentation

By fusing L2 directly into the base weights, you get a model that always applies the coding-specialist transformations without needing runtime LoRA switching or adapter loading.

🚀 Usage

Transformers

python

import torchfrom transformers import AutoModelForCausalLM, AutoTokenizer model_id = "petergilani/Macaron-V1-Tall-L2-Coding-Fused-BF16" tokenizer = AutoTokenizer.from_pretrained(    model_id,    trust_remote_code=True,) model = AutoModelForCausalLM.from_pretrained(    model_id,    torch_dtype=torch.bfloat16,    device_map="auto",    trust_remote_code=True,)model.eval() messages = [    {"role": "user", "content": "Write a Python function that performs a binary search on a sorted list and returns the index of the target."}] inputs = tokenizer.apply_chat_template(    messages,    add_generation_prompt=True,    return_tensors="pt",).to(model.device) generated_ids = model.generate(    **inputs,    max_new_tokens=512,    do_sample=True,    temperature=0.2,)print(tokenizer.decode(generated_ids[0][inputs["input_ids"].shape[-1]:], skip_special_tokens=True))

Converting to MLX

This PyTorch BF16 checkpoint can be converted to MLX format for Apple Silicon inference:

bash

python -m mlx_lm.convert \    --torch-sharded-path petergilani/Macaron-V1-Tall-L2-Coding-Fused-BF16 \    --mlx-path mlx-model/

Then run inference:

python

from mlx_lm import load, generate model, tokenizer = load("mlx-model/") prompt = "Implement a simple REST API with FastAPI that returns 'Hello, World!'"output = generate(model, tokenizer, prompt=prompt, verbose=True)

📜 License

This fused model inherits the MIT License from the original Macaron-V1-Tall repository.

Users should also respect any requirements inherited from the Qwen3.6-35B-A3B base model.

📚 Citation

If you use this model in your work, please cite the original Macaron-V1 paper:

bibtex

@misc{mindlab2026macaronv1,  author = {{Mind Lab}},  title = {Introducing Macaron-V1},  year = {2026},  howpublished = {Mind Lab: A Lab for Experiential Intelligence},  note = {https://macaron.im/mindlab/research/introducing-macaron-v1}}

And the base model:

bibtex

@misc{qwen3.6,  author = {Qwen Team},  title = {Qwen3.6-35B-A3B},  year = {2026},  publisher = {Alibaba},  url = {https://huggingface.co/Qwen/Qwen3.6-35B-A3B}}

Original model: mindlab-research/Macaron-V1-Tall
Base architecture: Qwen/Qwen3.6-35B-A3B
Macaron Artifacts: github.com/MindLab-Research/macaron-artifacts
MoL Serving Harness: github.com/MindLab-Research/Mixture-of-LoRA-Harness
Blog: Introducing Macaron-V1
Hosted API: Macaron API Platform

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Model Details

Model Provider

petergilani

Model Tree

Base

mindlab-research/Macaron-V1-Tall

Fine-tuned

this model

Input Modalities

TextImageVideo

Output Modalities

Text

Supported Functionality

Dedicated Endpoints

Explore FriendliAI today

Get started Talk to an engineer

🔍 What Was Fused

Table with columns: Component, Status
Component	Status
Base model	`Macaron-V1-Tall` (Qwen3.6-35B-A3B, BF16) ✅
L2 Coding LoRA (rank 64)	Fused via `W_fused = W_base + (B @ A)` ✅
L0 Chat LoRA	❌ Not included (remains an adapter)
L1 Agent LoRA	❌ Not included (remains an adapter)
L3 GenUI LoRA	❌ Not included (remains an adapter)
MTP module	❌ Removed (`mtp_num_hidden_layers = 0`)

Verification Details

Fusion was confirmed by inspecting model.safetensors.index.json:

No LoRA keys: The weight map contains only standard base model keys (linear_attn.*, self_attn.*, mlp.switch_mlp.*, mlp.shared_expert.*, mlp.gate.*). There are zero lora_A, lora_B, or adapter_* keys — confirming all L2 LoRA weights were absorbed into the base.
Parameter count: total_parameters = 34,660,608,768 (~34.7B) — matches the Macaron-V1-Tall base exactly, since a rank-64 LoRA adds negligible parameters that are now embedded in the base weights.
MTP stripped: mtp_num_hidden_layers = 0 confirms the Multi-Token Prediction module was removed as documented in the fusion plan.
Sharding: 14 safetensors files totaling ~64.5 GB at BF16 precision.

🏗️ Architecture Summary

Table with columns: Field, Value
Field	Value
Base model	Macaron-V1-Tall (Qwen3.6-35B-A3B)
Fusion method	Linear combination: `W_fused = W_base + (B @ A)`
LoRA rank	64
LoRA target	L2 Coding specialist
Context length	262K
Layers	40
Hidden size	2048
Attention heads	16

What the L2 Coding LoRA Adds

The L2 Coding specialist was trained for:

Repository-level coding — understanding project context, navigating codebases
SWE tasks — bug fixing, feature implementation, refactoring
Terminal use — shell commands, git operations, build/test workflows
Code-native workflows — tool calling, code generation, documentation

By fusing L2 directly into the base weights, you get a model that always applies the coding-specialist transformations without needing runtime LoRA switching or adapter loading.

🚀 Usage

Transformers

python

import torchfrom transformers import AutoModelForCausalLM, AutoTokenizer model_id = "petergilani/Macaron-V1-Tall-L2-Coding-Fused-BF16" tokenizer = AutoTokenizer.from_pretrained(    model_id,    trust_remote_code=True,) model = AutoModelForCausalLM.from_pretrained(    model_id,    torch_dtype=torch.bfloat16,    device_map="auto",    trust_remote_code=True,)model.eval() messages = [    {"role": "user", "content": "Write a Python function that performs a binary search on a sorted list and returns the index of the target."}] inputs = tokenizer.apply_chat_template(    messages,    add_generation_prompt=True,    return_tensors="pt",).to(model.device) generated_ids = model.generate(    **inputs,    max_new_tokens=512,    do_sample=True,    temperature=0.2,)print(tokenizer.decode(generated_ids[0][inputs["input_ids"].shape[-1]:], skip_special_tokens=True))

Converting to MLX

This PyTorch BF16 checkpoint can be converted to MLX format for Apple Silicon inference:

bash

python -m mlx_lm.convert \    --torch-sharded-path petergilani/Macaron-V1-Tall-L2-Coding-Fused-BF16 \    --mlx-path mlx-model/

python -m mlx_lm.convert \    --torch-sharded-path petergilani/Macaron-V1-Tall-L2-Coding-Fused-BF16 \    --mlx-path mlx-model/

Then run inference:

python

from mlx_lm import load, generate model, tokenizer = load("mlx-model/") prompt = "Implement a simple REST API with FastAPI that returns 'Hello, World!'"output = generate(model, tokenizer, prompt=prompt, verbose=True)

📜 License

This fused model inherits the MIT License from the original Macaron-V1-Tall repository.

Users should also respect any requirements inherited from the Qwen3.6-35B-A3B base model.

📚 Citation

If you use this model in your work, please cite the original Macaron-V1 paper:

bibtex

@misc{mindlab2026macaronv1,  author = {{Mind Lab}},  title = {Introducing Macaron-V1},  year = {2026},  howpublished = {Mind Lab: A Lab for Experiential Intelligence},  note = {https://macaron.im/mindlab/research/introducing-macaron-v1}}

@misc{mindlab2026macaronv1,  author = {{Mind Lab}},  title = {Introducing Macaron-V1},  year = {2026},  howpublished = {Mind Lab: A Lab for Experiential Intelligence},  note = {https://macaron.im/mindlab/research/introducing-macaron-v1}}

And the base model:

bibtex

@misc{qwen3.6,  author = {Qwen Team},  title = {Qwen3.6-35B-A3B},  year = {2026},  publisher = {Alibaba},  url = {https://huggingface.co/Qwen/Qwen3.6-35B-A3B}}

@misc{qwen3.6,  author = {Qwen Team},  title = {Qwen3.6-35B-A3B},  year = {2026},  publisher = {Alibaba},  url = {https://huggingface.co/Qwen/Qwen3.6-35B-A3B}}

Original model: mindlab-research/Macaron-V1-Tall
Base architecture: Qwen/Qwen3.6-35B-A3B
Macaron Artifacts: github.com/MindLab-Research/macaron-artifacts
MoL Serving Harness: github.com/MindLab-Research/Mixture-of-LoRA-Harness
Blog: Introducing Macaron-V1
Hosted API: Macaron API Platform

Macaron-V1-Tall-L2-Coding-Fused-BF16

README

🔍 What Was Fused

Verification Details

🏗️ Architecture Summary

What the L2 Coding LoRA Adds

🚀 Usage

Transformers

Converting to MLX

📜 License

📚 Citation

Explore FriendliAI today

README

🔍 What Was Fused

Verification Details

🏗️ Architecture Summary

What the L2 Coding LoRA Adds

🚀 Usage

Transformers

Converting to MLX

📜 License

📚 Citation

Macaron-V1-Tall-L2-Coding-Fused-BF16

README

🔍 What Was Fused

Verification Details

🏗️ Architecture Summary

What the L2 Coding LoRA Adds

🚀 Usage

Transformers

Converting to MLX

📜 License

📚 Citation

🔗 Related Links

Explore FriendliAI today

README

🔍 What Was Fused

Verification Details

🏗️ Architecture Summary

What the L2 Coding LoRA Adds

🚀 Usage

Transformers

Converting to MLX

📜 License

📚 Citation

🔗 Related Links