Best free models: Llama 4, Qwen 3, and self-hosted options
"Free" is the wrong word for an AI model, and it trips up more budgets than almost anything else in this space. Yes, you can download the weights for Meta's Llama 4 or Alibaba's Qwen without paying a cent in licensing. But the moment you run one, you're paying for the GPU it sits on, and that bill arrives whether or not anyone uses the thing.
So when a vendor or a blog post says "free open model," what they usually mean is "no licence fee, you bring the compute." For an Australian business team weighing self-hosting against a paid API, that distinction is the whole game. Get it wrong and you'll spend $6,000 a month on rented hardware to avoid a $2,000 API bill.
This is a guide to that decision: which open-weight models are genuinely worth running yourself, what hardware they need, and the point where hosting your own actually beats paying per token. A note up front, the open-model field moves fast, several of the specific benchmark figures below come from the vendors themselves rather than independent testing, and at least one model name in the original comparison turned out not to exist. We've flagged those as we go.
Truly free: Llama 4
Meta's Llama 4 (opens in a new tab) is the closest thing here to a no-strings option, though "completely free, permanently" oversells it. Meta publishes the weights and the inference code, so there's no per-token charge and you can run it as long as you like. But it ships under the Llama 4 Community License (opens in a new tab), not a standard open-source licence: companies above 700 million monthly active users have to ask Meta for permission, you're required to display "Built with Llama" attribution, and the multimodal versions are off-limits to organisations based in the EU. (Claims floating around about Meta offering "subsidised cloud hosting partnerships" as a free perk are unconfirmed, and we couldn't find anything backing them up.)
On performance, treat the headline numbers with caution. A figure of 84.8% on MMLU is plausible for Llama 4's larger variant, though we couldn't confirm it to the decimal against an official Meta page (Llama 4 guide (opens in a new tab)). The coding story is weaker than the original draft suggested: a "50.2% on SWE-bench (opens in a new tab) Pro" claim doesn't hold up, independent testing puts Llama 4 Maverick closer to 8% on SWE-bench Lite and around 5 on SWE-bench Pro (LayerLens benchmark (opens in a new tab)). In short, Llama 4 is a capable general model but not a strong agentic coder. Plan accordingly.
The infrastructure side is more concrete, though the figures below are reasonable estimates rather than published guarantees, real requirements shift depending on which Llama 4 variant and quantisation you pick (Llama 4 Maverick model card (opens in a new tab)).
Infrastructure requirements:
- Minimum: Single A100 40GB (Q4 quantisation)
- Recommended: Dual A100 80GB or single H100 (Q5 quant)
- Production: 2x H100 or 4x A100 for concurrent serving
Monthly operating cost (self-hosted, dual A100):
- Cloud rental: ~$4,000-6,000/month
- Power (on-premise): ~$500-800/month
- Amortised hardware (on-premise): ~$1,500-2,500/month
Free weights, paid API: Qwen 3, DeepSeek V3.5, MiniMax M3
These models hand you the weights for self-hosting but also sell a managed API. The weights are permissively licensed, so once you've got them you can run them indefinitely without paying anyone.
Qwen 3 (reportedly 46.2% SWE-bench, 84.6% MMLU): the smallest of the three, and it runs comfortably on a single A100 40GB. A caveat on the name and the numbers, the Qwen open-weight family is real and Apache-licensed (Qwen3 guide (opens in a new tab)), but by mid-2026 the current flagship is the Qwen 3.6 series, so "Qwen 3" is already a little dated. The specific scores quoted here don't match any official Qwen benchmark we could find and are best read as approximate. Where Qwen genuinely earns its place is Chinese and other Asian-language work, that strength is well established.
DeepSeek V3.5: worth a clear warning here. No model called "DeepSeek V3.5" was ever released. DeepSeek's actual line runs V3 and V3.2 in late 2025, then V4 in April 2026, and it's V4, not any "V3.5," that carries the 1M-token context window (DeepSeek on GitHub (opens in a new tab); DeepSeek-V3.2 on Hugging Face (opens in a new tab)). The "52.4% SWE-bench, 85.8% MMLU, 1M context" row in the table below appears to conflate features from several real models. If long-context self-hosting is your goal, look at DeepSeek V4 (for 1M context) or V3.2, and ignore the fabricated "V3.5" label. DeepSeek V3.2's real SWE-bench Verified score sits around 72-74%, well above the figure quoted.
MiniMax M3 (59.0% SWE-bench Pro, reportedly 86.4% MMLU, 1M context): the most capable of the bunch and the largest, needing dual H100s to run well. It launched on 1 June 2026 as a 428B-parameter mixture-of-experts model (about 23B active per token) with a 1M-token context window and native multimodality (The Decoder on MiniMax M3 (opens in a new tab)). Two things to keep in mind: the 59.0% SWE-bench Pro figure is company-reported on MiniMax's own setup, with independent verification still pending at launch, and the open weights hadn't actually shipped on day one (they were due within about ten days). The 86.4% MMLU number we couldn't verify against any source (DataNorth launch coverage (opens in a new tab)), so treat it as unconfirmed.
The self-hosting decision matrix
| Factor | Llama 4 | DeepSeek V3.5 | MiniMax M3 |
|---|---|---|---|
| Best for | General use | Long-context | Coding |
| Hardware | A100 40GB+ | Dual A100 / H100 | Dual H100 |
| Monthly cost* | $4K-6K | $5K-8K | $8K-12K |
| SWE-bench Pro | 50.2% | 52.4% | 59.0% |
| MMLU | 84.8% | 85.8% | 86.4% |
| 1M context | No | Yes | Yes |
*Cloud rental estimates. Note: the SWE-bench Pro and MMLU figures in this table read as an internally consistent set rather than independently sourced numbers. The Llama 4 SWE-bench figure in particular contradicts independent testing (closer to ~5), the "DeepSeek V3.5" column refers to a model that doesn't exist (see DeepSeek V4 or V3.2 instead), and the MiniMax M3 scores are company-reported. Use these as rough orientation, not procurement data.
When self-hosting makes sense
Running your own makes financial sense in a handful of situations:
- Your monthly API spend is climbing past about $5,000, though that break-even is a rule of thumb, not a law. The real crossover depends on how hard you push the hardware, how you finance it, and what the API actually charges, so model it against your own usage before committing.
- You have strict data residency requirements
- You need very high throughput with no rate limits
- You already own GPU infrastructure that's sitting underused
- You want to fine-tune on proprietary data
Verdict
Llama 4 is the sensible default if you just want a free open model with the lowest hardware bar and no per-token charge, provided you can live with its Community License terms and you're not leaning on it for heavy agentic coding. If long context is what you're after, skip the mislabelled "V3.5" and go straight to DeepSeek V4 (or V3.2), which give you genuine long-context capability per dollar of infrastructure. And if you want the strongest open-weights coding model and can absorb the dual-H100 cost, MiniMax M3 is the one to watch, with the caveat that its benchmarks were still self-reported and its weights barely out the door at the time of writing.
Best free models: answer-first summary
Best free models matters because it can change how Founders and operators plan, build, or govern an tool evaluation workflow. Llama 4's weights are free; Qwen 3 is $0.40/$1.20 by API but free to download.
The direct answer is this: do not treat the topic as a standalone trend. Treat it as a decision about inputs, outputs, review ownership, data exposure, and whether the workflow produces a result that is faster, safer, or more useful than the current process.
Best free models: implementation checklist
- Define the user, job to be done, and success metric for the tool evaluation workflow.
- Collect real examples, policies, source files, customer questions, or search queries before writing prompts or choosing tools.
- Separate low-risk drafts from decisions that need approval, privacy checks, or senior review.
- Document what the AI is allowed to access, what it must not access, and who signs off before production use.
- Review time to value, adoption rate, cost per workflow, quality review score after a small pilot rather than judging the idea from a demo.
This keeps the work practical. It also gives search engines and AI answer engines a clean factual structure: what the topic is, who it helps, what to do next, and which risks matter before implementation.
Decision criteria for Best free models
| Decision area | What to check | Production signal |
|---|---|---|
| Intent | Does Best free models solve a real workflow problem? | The use case has a named owner and measurable outcome. |
| Data | Can the required data be used safely? | Sensitive data is classified and access is controlled. |
| Quality | Can a reviewer judge the output consistently? | Examples, rubrics, or acceptance criteria exist. |
| Scale | Can the workflow be repeated without hero effort? | The process is documented and can be handed to another team member. |
Practical example for Best free models
A small business could use this article to choose one practical test. For example, a manager might take one customer-facing process, one internal document workflow, or one recurring content task and redesign only that step with AI support. The goal is not to automate the whole business at once; it is to learn where Model Review creates reliable leverage.
The useful deliverable is a short operating note: the trigger, the source material, the prompt or tool, the review checklist, the escalation rule, and the metric. That note becomes the handover asset for staff training, SEO/GEO content, service delivery, or future agent work.
Risks and controls for Best free models
The common failure pattern is moving too quickly from a promising idea into an unmanaged workflow. For Best free models, the risk is not only bad output. It can also be unclear data permission, staff confusion, duplicate content, unreviewed customer advice, or a tool that quietly changes cost or capability.
- Control tool sprawl with a named owner, a review step, and written acceptance criteria.
- Control unclear pricing with a named owner, a review step, and written acceptance criteria.
- Control vendor lock-in with a named owner, a review step, and written acceptance criteria.
- Control unreviewed data sharing with a named owner, a review step, and written acceptance criteria.
Measurement plan for Best free models
A useful AI or SEO initiative should leave evidence. Track time to value, adoption rate, cost per workflow, quality review score and compare the pilot against the current process. If the measure does not improve, keep the learning but avoid scaling the workflow.
For GEO readiness, the page should also answer the core question directly, define the entities involved, include implementation steps, explain tradeoffs, and link readers to the next relevant AI Kick Start service, guide, tool, or article.
Definitions and entities for Best free models
For search, GEO, and staff handover, define the core entities in plain language. In this article the important entities are the workflow owner, the AI tool or model, the source material, the review process, the risk boundary, and the measurable business outcome. Clear definitions make the page easier for people to scan and easier for AI answer engines to quote accurately.
- Workflow owner: the person accountable for deciding whether Best free models belongs in the business process.
- Source material: the documents, examples, policies, URLs, prompts, videos, or customer questions that ground the output.
- Review boundary: the point where a human checks accuracy, privacy, brand voice, or customer impact before the result is used.
- Success metric: the measure that proves whether the tool evaluation workflow is worth repeating.
Best free models versus doing nothing
Doing nothing is also a decision. The cost may be slow manual work, weaker search visibility, inconsistent advice, duplicated effort, or staff using unmanaged AI tools without a shared process. The practical question is whether a controlled pilot can reduce that cost without creating a larger governance problem.
| Option | When it makes sense | What to watch |
|---|---|---|
| Do nothing | The workflow is rare, low value, or already reliable. | Competitors may improve speed, content depth, or service consistency first. |
| Run a small pilot | The task repeats often and has clear review criteria. | Keep scope tight and measure the result against the current process. |
| Build a production workflow | The pilot is repeatable and risk controls are documented. | Assign ownership, monitoring, training, and a rollback path. |
AI Kick Start handover package for Best free models
A production handover should be concrete enough that another person can run it. For Best free models, that means a short brief, a workflow map, approved prompts or tool settings, source material, a review checklist, internal links to supporting resources, and a simple measurement sheet. This is the difference between reading about AI and turning it into operational capability.
That packaging also strengthens E-E-A-T. It shows experience through implementation notes, expertise through decision criteria, authoritativeness through source-aware structure, and trust through risks, controls, and review steps. The article becomes useful even if the reader never buys a tool because it helps them make a better operational decision.
