# LFM3

Hacker Approved page: https://hackerapproved.com/p/lfm3/

7 people recommended LFM3 in Hacker News comments, 3 had mixed feelings and 2 criticized it, November 2025 to October 2026. It was recommended in 1 of 12 months, in 1 thread. It does not meet the Hacker Approved badge rule.

7 of the 7 people who recommend it commented in one thread, “Introducing LFM3: Every Modality for Specialized Intelligence”.

**Verdict:** People recommend the small LFM3 models for strong text ability relative to their size and for running on modest memory. Critics say its coding output is weak and its answers are sometimes inaccurate.

Summary written by Claude from 15 comments.

## Why people recommend it

- Strong for its size (4 comments in the sample): "It's a model designed to be tiny and still functional, unlike other larger models that are just scaled down" ([source](https://news.ycombinator.com/item?id=50265273)) "LFM3-1.7B is better at pure text tasks than all other models" ([source](https://news.ycombinator.com/item?id=50268047)) "I think the d1-2B is amazing, it can run on 1GB of unified RAM." ([source](https://news.ycombinator.com/item?id=50263207))
- Step up from earlier versions (2 comments in the sample): "Excited to roll out LFM 3 to replace it, looks like a big step up." ([source](https://news.ycombinator.com/item?id=50263408)) "to beat Qwen3.5-4B with such a capable model is amazing" ([source](https://news.ycombinator.com/item?id=50267126))
- Runs in about 1GB memory (1 comment in the sample): "LFM3-1.7B fits in about 1 GB at 4-bit, holds 256K context" ([source](https://news.ycombinator.com/item?id=50263584))

## What people warn about

- Weak at coding, spaghetti code (2 comments in the sample): "definitely would not recommend using it for SWE-style tasks. But chaining a couple bash calls together to find a file? no problem" ([source](https://news.ycombinator.com/item?id=50268077)) "It will produce code that mixes too many things at once, creating spaghetti code" ([source](https://news.ycombinator.com/item?id=50285530))

## Other small language models people recommend

- [OpenAI Luna](https://hackerapproved.com/p/luna/): 171 recommend, 31 mixed, 46 criticize
- [Claude Haiku](https://hackerapproved.com/p/claude-haiku/): 147 recommend, 48 mixed, 113 criticize
- [Apple Foundation Models](https://hackerapproved.com/p/apple-foundation-models/): 24 recommend, 17 mixed, 31 criticize
- [Ministral](https://hackerapproved.com/p/ministral/): 14 recommend, 3 mixed, 5 criticize

## The badge

A product earns the Hacker Approved badge when at least 10 people recommend it, recommenders outnumber critics at least 2 to 1 (mixed feelings count toward neither side), it was recommended in at least 4 of the 12 months and in at least 5 different threads, and no single thread supplies more than half of the people who recommend it. Health and money products are never badged.

## Cite as

Hacker Approved, "LFM3", https://hackerapproved.com/p/lfm3/ (7 recommend, 3 mixed, 2 criticize; Hacker News comments, November 2025 to October 2026).

Counts updated 2 Nov 2026, with data through October 31, 2026.