---
license: apache-2.0
base_model: Qwen/Qwen-AgentWorld-35B-A3B
library_name: mlx
tags:
- mlx
- omlx
- lm-studio
- qwen
- tool-calling
---

# jedisct1/Qwen-AgentWorld-35B-A3B-oQ6-MLX

This is an oMLX MLX quantization of `Qwen/Qwen-AgentWorld-35B-A3B`.

The goal of this package is practical local agent use on Apple Silicon. Tool calling was treated as the release gate, so the chat template is kept both as `chat_template.jinja` and embedded in `tokenizer_config.json`, with `tool_parser_type` set to `qwen3_coder`.

## Quantization

- Quantization: oQ6 (6-bit affine, mixed precision)
- Global bits: 6
- Group size: 64
- Mode: affine
- Safetensor shards: 6
- Tensor count: 1677
- Total safetensor size: 28.51 GB

The upstream config declares an MTP layer, but the upstream checkpoint published for `Qwen/Qwen-AgentWorld-35B-A3B` does not include `mtp.*` tensors. These artifacts therefore publish a self-consistent non-MTP config instead of advertising a missing draft head.

## Compatibility

Expected local targets:

- oMLX 0.4.4 or newer
- LM Studio with MLX model loading and OpenAI-compatible tool calls

Use greedy decoding for strict tool use and eval runs:

```json
{"temperature": 0, "top_p": 1}
```

## Tool-Calling Verification

Swival core tool suite: 5/5 passed on 2026-06-24 with deterministic greedy decoding.

Swival all-built-ins suite: 5/5 passed on 2026-06-24 with deterministic greedy decoding.

The Swival suites exercise real file, edit, command, planning, checklist, snapshot, grep, outline, URL fetch, batched reads, and shell-tool dispatch through an OpenAI-compatible server. A direct `/v1/chat/completions` smoke also checks that the server returns structured `tool_calls`, not just plain XML text.