Hayz Labs · https://hayzlabs.com/reviews/best/open-source/
reviews · open source
The best open source AI for each job
The best-scored option whose code or model weights are published, so you can inspect it, run it yourself and keep using it if the maker changes course.
New to this? What the scores and labels mean
- 0 to 100
- Our score for one job, like writing or coding, compared with everything else out today. 100 is the best available today. For AI models we build it from public data, such as Arena's blind votes and Epoch AI's independent tests; apps and tools are scored on our own checklist. The same tool can score differently for different jobs, and where two likely ranges overlap, we call it a tie.
- Our verdict
- Use it: a top choice for most people. Worth trying: good, and worth a go. Depends: good for some people or one kind of job. Wait: too early to tell, or not out for everyone yet. Skip: something else does the job better.
- First look
- Just released. We say what it is and whether to bother, and give it a score once public results exist.
- No score yet
- None of the public sources we use covers it yet, or the only evidence is from a source whose numbers we can't use, so we describe it in words instead.
- Best buy
- Nothing cheaper scores as well for that job.
- Best free
- The best score for that job that costs nothing, through a free plan or by running on your own computer.
- Pay as you go
- No monthly plan: you pay for each use, usually through a developer account.
- Checked
- The day we last confirmed the score and the price against the sources. Prices change often, so check the date.
- AI model
- The engine underneath, like GPT or Claude. You use it through an app or, if you code, directly.
- Open source
- Free to download and run on your own computer, so nothing leaves your machine.
An answer for 7 of 8 jobs. Where nothing we have scored fits, we say so rather than stretch the rule. 100 is the best available today in that job.
Some picks here are made by Anthropic, whose Claude models help draft our reviews. Each pick also names the best option from another maker.
Writing
Wait Mistral's 1T preview scores no higher than 72 in any job; weights due 31 Oct. Use GLM-5.3 or Kimi K3 now.
Clearly better, if you can drop that rule: Claude Opus 5.5 scores 100 against 59 here, for $20 a month. Worth it if this job is most of your work.
Research and long documents
Worth trying Top open-weights model on the Artificial Analysis index, at cents per task; closed leaders still rank well above it.
Close to #1 (Gemini 4 Argon, 95), so there is little to gain by looking further.
From another maker: GLM-5.3-Flash (74, free)
Coding
Worth trying Free, private way to run open coding models like Qwen3.8 27B on your own computer; skip it if you need frontier quality.
Clearly better, if you can drop that rule: Claude Opus 5.5 scores 100 against 80 here, for $20 a month. Worth it if this job is most of your work.
From another maker: OpenCode (80, free)
Data and spreadsheets
Worth trying Very cheap MIT model, 13th on arena.ai Math; for coding and agents MiMo-V2.6-Pro ranks higher for less per task.
Close to #1 (Gemini 3.8 Flash, 85), so there is little to gain by looking further.
Images
Depends Best scoring image model you can download and run yourself, but its license bars commercial use. Fine for research.
Clearly better, if you can drop that rule: GPT Image 2.5 Sunburst scores 100 against 48 here, for pay as you go. Worth it if this job is most of your work.
Voice and audio
Nothing we have scored for voice and audio fits yet. See the full ranking.
Video
Use it Best value in video: top-rated image to video with sound at about $5 a minute. Weights barred in US, EU, UK.
Clearly better, if you can drop that rule: Gemini Omni Flash 1.1 scores 100 against 84 here, for $4.99 a month. Worth it if this job is most of your work.
Agents
Worth trying Top open-weights model on the Artificial Analysis index, at cents per task; closed leaders still rank well above it.
Close to #1 (Claude Opus 5.5, #1 on Arena's Agent board), so there is little to gain by looking further.
From another maker: Mistral Large 4 (#39 on Arena's Agent board, pay as you go)