# Swift-1.5-Qwen3.8-27b-oQ8e-mtp achieves 34.8 tokens per second on Apple M5 Max

Digest AI · Research · published 2026-09-29T16:07:22Z

Canonical: https://digestai.news/story/swift-1-5-qwen3-8-27b-oq8e-mtp-achieves-34-8-tokens-per-second-on-appl

## Summary

A Reddit user shared benchmark results for the **Swift-1.5-Qwen3.8-27b-oQ8e-mtp** model on an **Apple M5 Max** device on September 29, 2026. The model reached **34.8 tokens per second** in a coding scenario, generating a playable game in a single HTML file without edits. The benchmark tool and parameters are detailed for reproducibility, though no official confirmation or context from **Qwen** or **Apple** is provided.

## Key points

- Swift-1.5-Qwen3.8-27b-oQ8e-mtp hit **34.8 tokens per second** on Apple M5 Max in a coding task
- Median speed across five runs was **34.5 tokens per second**, with **KV cache** settings affecting results

## Why it matters

Faster inference on Apple silicon could lower costs for developers running large models locally, but benchmarks from community sources lack official validation.

## Sources

1. [Swift-1.5-Qwen3.8-27b-oQ8e-mtp on Apple M5 Max — 34.8 tok/s — llm-bench.io](https://llm-bench.io/benchmarks/cmumgubnk004o01o00kqgaqcv) (llm-bench.io, 2026-09-29)

Part of the developing story: [Local AI Inference Speeds Hit New Peaks](https://digestai.news/thread/llama-cpp-achieves-up-to-42x-faster-prompt-lookup-drafting) (2 stories)

## Cite

Digest AI, "Swift-1.5-Qwen3.8-27b-oQ8e-mtp achieves 34.8 tokens per second on Apple M5 Max", 29 September 2026, https://digestai.news/story/swift-1-5-qwen3-8-27b-oq8e-mtp-achieves-34-8-tokens-per-second-on-appl

---

Written by Digest AI's editorial model from the linked sources; the sources are the record. Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse
JSON: https://digestai.news/story/swift-1-5-qwen3-8-27b-oq8e-mtp-achieves-34-8-tokens-per-second-on-appl.json