batiai/ qwen3.5-27b:iq4

64 Downloads Updated 3 weeks ago

tools thinking

ollama run batiai/qwen3.5-27b:iq4

curl http://localhost:11434/api/chat \
  -d '{
    "model": "batiai/qwen3.5-27b:iq4",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='batiai/qwen3.5-27b:iq4',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'batiai/qwen3.5-27b:iq4',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 3 weeks ago

3 weeks ago

c21796f3889c · 15GB ·

model

archqwen35

·

parameters26.9B

·

quantizationIQ4_XS

15GB

template

{{- $lastUserIdx := -1 -}} {{- range $idx, $msg := .Messages -}} {{- if eq $msg.Role "user" }}{{ $la

1.7kB

system

You are a helpful AI assistant.

31B

params

{ "num_ctx": 131072, "stop": [ "<|im_end|>", "<|endoftext|>", "<|im_

144B

Readme

Qwen 3.5 27B — Quantized by BatiAI

Quantized from official Alibaba weights. Verified on real Mac hardware.

Models

Tag	Size	VRAM	M4 Max (128GB)	Use Case
iq4	14GB	28GB	17.0 t/s	32GB+ Mac

Quick Start

ollama run batiai/qwen3.5-27b:iq4

Why Qwen 3.5 27B?

Hybrid architecture (Gated DeltaNet + GQA + MoE)
262K context window
Excellent Korean + tool calling + coding
Apache 2.0 license

RAM Requirements

Your Mac RAM	IQ4 (14GB)
16GB	❌
32GB	✅ Fits (28GB VRAM)
48GB+	✅ Fast
128GB	17.0 t/s

Comparison

Model	Size	VRAM	Speed (M4 Max)
batiai/qwen3.5-9b:q4	5.2GB	—	12.5 t/s (16GB Mac)
batiai/qwen3.5-27b:iq4	14GB	28GB	17.0 t/s
batiai/qwen3.5-35b:iq4	17GB	23GB	26.6 t/s

For 32GB Mac, 27B is the best Qwen option. For 36GB+, the 35B MoE is faster.

Why BatiAI?

Quantized directly from official Alibaba weights
IQ4_XS with imatrix — best quality at this size
Verified on MacBook Pro M4 Max (128GB)
Korean, tool calling, JSON generation all tested

Built for BatiFlow

Free, on-device AI automation for Mac. 5MB app, 100% local, unlimited.

https://flow.bati.ai

# Qwen 3.5 27B — Quantized by BatiAI

Quantized from official Alibaba weights. Verified on real Mac hardware.

## Models

| Tag | Size | VRAM | M4 Max (128GB) | Use Case |
|-----|------|------|----------------|----------|
| **iq4** | **14GB** | **28GB** | **17.0 t/s** | **32GB+ Mac** |

## Quick Start

```
ollama run batiai/qwen3.5-27b:iq4
```

## Why Qwen 3.5 27B?

- Hybrid architecture (Gated DeltaNet + GQA + MoE)
- 262K context window
- Excellent Korean + tool calling + coding
- Apache 2.0 license

## RAM Requirements

| Your Mac RAM | IQ4 (14GB) |
|-------------|-----------|
| 16GB | ❌ |
| 32GB | ✅ Fits (28GB VRAM) |
| 48GB+ | ✅ Fast |
| 128GB | 17.0 t/s |

## Comparison

| Model | Size | VRAM | Speed (M4 Max) |
|-------|------|------|----------------|
| batiai/qwen3.5-9b:q4 | 5.2GB | — | 12.5 t/s (16GB Mac) |
| **batiai/qwen3.5-27b:iq4** | **14GB** | **28GB** | **17.0 t/s** |
| batiai/qwen3.5-35b:iq4 | 17GB | 23GB | 26.6 t/s |

For 32GB Mac, 27B is the best Qwen option. For 36GB+, the 35B MoE is faster.

## Why BatiAI?

- Quantized directly from official Alibaba weights
- IQ4_XS with imatrix — best quality at this size
- Verified on MacBook Pro M4 Max (128GB)
- Korean, tool calling, JSON generation all tested

## Built for BatiFlow

Free, on-device AI automation for Mac. 5MB app, 100% local, unlimited.

https://flow.bati.ai

Paste, drop or click to upload images (.png, .jpeg, .jpg, .svg, .gif)