Swift-Qwen3.8-27B
Shares rank 66 of 72 language models by recommenders
Cuts thinking tokens by roughly 40-60% while keeping accuracy close to the base model, and feels usable day to day. Critics note it scores somewhat worse on benchmarks.
Cuts thinking tokens by roughly 40-60% while keeping accuracy close to the base model, and feels usable day to day. Critics note it scores somewhat worse on benchmarks.
7 people gave an opinion
- recommend it
- 6
- criticize it
- 1
- approval share
- 86%
Recommended in 2 of 13 months
- November 2025: not recommended
- December 2025: not recommended
- January 2026: not recommended
- February 2026: not recommended
- March 2026: not recommended
- April 2026: not recommended
- May 2026: not recommended
- June 2026: not recommended
- July 2026: not recommended
- August 2026: not recommended
- September 2026: recommended
- October 2026: recommended
- November 2026: not recommended (month in progress)
Why it doesn’t carry the badge
- 6 people recommend itNeeds 10 or more, not met.
- 6 recommenders for every criticNeeds 2 to 1 or better, met.
- Recommended in 2 of 13 monthsNeeds 4 or more, not met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Cuts thinking tokens substantially
“It certainly seems to cut down on thinking compared to stock Qwen3.8 27B in my (limited) testing.”
Comment from 16 Sep 2026 in “Show HN: Swift-Qwen3.8-27B, -58.3% thinking, x1.95 speed, accuracy of xhigh” “We got almost no loss (we got Base 66.74% vs Swift 65.84%) with -38.7% thinking token reduction.”
Comment from 16 Sep 2026 in “Show HN: Swift-Qwen3.8-27B, -58.3% thinking, x1.95 speed, accuracy of xhigh” “performance converged to 40-60% token reduction with <1% accuracy loss”
Comment from 16 Sep 2026 in “Show HN: Swift-Qwen3.8-27B, -58.3% thinking, x1.95 speed, accuracy of xhigh” Little accuracy loss versus base
“We got almost no loss (we got Base 66.74% vs Swift 65.84%) with -38.7% thinking token reduction.”
Comment from 16 Sep 2026 in “Show HN: Swift-Qwen3.8-27B, -58.3% thinking, x1.95 speed, accuracy of xhigh” “performance converged to 40-60% token reduction with <1% accuracy loss”
Comment from 16 Sep 2026 in “Show HN: Swift-Qwen3.8-27B, -58.3% thinking, x1.95 speed, accuracy of xhigh” Usable for day-to-day work
“very rarely (as opposed to the Base Qwen 3.8 27B), making it actually usable for day-to-day work”
Comment from 17 Sep 2026 in “Show HN: Swift-Qwen3.8-27B, -58.3% thinking, x1.95 speed, accuracy of xhigh” “it's quite amazing for the amount I've used it”
Comment from 16 Sep 2026 in “Show HN: Swift-Qwen3.8-27B, -58.3% thinking, x1.95 speed, accuracy of xhigh”
What people warn about
Each mark is one comment
Worse on benchmarks
“Swift-Qwen is faster, but also - I understand - worse on benchmarks”
Comment from 6 Oct 2026 on Hacker News
Other language models people recommend
The four most recommended other language models, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Criticize | Approval | Months | Owned | Badge |
|---|---|---|---|---|---|---|
| Claude Opus | 3,331 | 2,746 | 55% | 13 | 4 mo156 owners | — |
| OpenAI GPT | 1,205 | 987 | 55% | 13 | 6 mo47 owners | — |
| Claude Sonnet | 854 | 586 | 59% | 13 | 6 mo46 owners | — |
| DeepSeek V4 Flash | 828 | 223 | 79% | 8 | 1 mo24 owners | Approved |
| Swift-Qwen3.8-27BThis page | 6 | 1 | 86% | 2 | — | — |
What people said
7 of the 9 comments that mention it, newest first. Each links to the original comment.
“I didn't find these any good either”
“Swift-Qwen is faster, but also - I understand - worse on benchmarks”
“very rarely (as opposed to the Base Qwen 3.8 27B), making it actually usable for day-to-day work”
“It certainly seems to cut down on thinking compared to stock Qwen3.8 27B in my (limited) testing.”
“We got almost no loss (we got Base 66.74% vs Swift 65.84%) with -38.7% thinking token reduction.”
“it's quite amazing for the amount I've used it”
“performance converged to 40-60% token reduction with <1% accuracy loss”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.