Infrastructure & DevOps
Solheim
Last updated 2026-08-15 · benchmark measured 2026-10-11 — deterministic & reproducible
EU-hosted LLM inference on a flat monthly fee, no token metering.
Solheim at a glance
Solheim provides private language model instances hosted on EU hardware for a flat monthly fee, aimed at developers running coding agents and founders who need a predictable inference backend.
Good for
- Developers running coding agents who need multiple concurrent requests (e.g., an agent, an editor, and a test loop running at once)
- Founders who need a private LLM backend for their product with a bill that doesn't change with traffic
- Teams that require EU-hosted, GDPR-compliant inference with no data leaving the region
Pricing
€15.00/month for 1 instance × 64k context; scales with instance count and context window (64k/128k/256k); dedicated/enterprise pricing beyond that.
Questions people ask
How does Solheim's pricing work?
You pay a flat monthly rate based on two settings: how many concurrent instances (requests in flight) you run and how large a context window each gets. There is no token counting, usage tiers, or overage charges.
What happens if I exceed my instance count?
Requests beyond your instance count queue up rather than failing with an error, and they do not cost extra.
Is Solheim compatible with my existing tools?
Yes, it uses an OpenAI-compatible API, so tools like Cline, ZooCode, VS Code BYOK, and any OpenAI SDK can connect without modification.
Where is my data hosted, and is it GDPR compliant?
Solheim runs on EU-owned GPU providers in EU datacenters, with no US hyperscaler in the serving path, giving a GDPR and AI Act compliant posture.
What models can I use?
The available open-weight models are Qwen3.6-35B-A3B, Qwen3.8-27B, and DeepSeek V4 Flash, each with up to 256k context, and you can request additional models.
Written by Legit.Show from solheim.ai on 2026-10-21. It does not change the score.
Are you the maker of Solheim? Claim this listing. It is free, takes a meta tag, a DNS record or GitHub admin rights, and never changes the score.
Launched something? Add your product, free.
To cite this score: legit.show/s/solheim-ai/2026-10-11. That address never changes; this page moves with every re-measure.
Legit.Show scored Solheim 81/100 on 2026-10-11, measured across all 7 frames from its public surface. legit.show/s/solheim-ai/2026-10-11
Is Solheim production-ready?
Legit.Show scores Solheim 81 out of 100 — the simple average of its 7 measured frames. Legit.Show ran its deterministic 7-Frame production-readiness benchmark on Solheim (public-surface assessment), measured from the public surface with no LLM in the scoring path. Its strongest frame is Performance; its weakest is Security. Every frame it averages is published with its evidence on the Legit.Show listing.
The 7 Frames
- Performance — 100/100
- Accessibility — 97/100
- Security — 25/100How this compares across the catalogue: The Web Security Baseline
- Privacy — 100/100How this compares across the catalogue: The Privacy Gap
- Reliability — 100/100
- Standards — 92/100
- Discoverability — 50/100
What we measured
- No Content-Security-Policy and no HSTS.
- Served over HTTPS with a valid certificate.
- Real Lighthouse performance run — 1024 ms to first byte.
- Returns a proper 404 for unknown routes.
- 3 of 3 sampled routes reachable.
- Has a reachable privacy policy.
- Sets cookies / loads scripts with no consent prompt.
- Discoverable: canonical URL.
Who it's for
Developers running coding agents · Founders needing inference backends · Teams requiring private LLMs · Products with predictable usage
Pricing
€15.00/month for 1 instance × 64k context; pricing scales by instance count and context window size; dedicated/enterprise options available
Sources and updates
- Description
- Taken from solheim.ai's own website on 2026-08-15.
- Operator
- Solheim OÜ, as named in solheim.ai's terms, privacy page or footer.
- Benchmark
- Measured by Legit.Show from the live site, the way any visitor sees it — with no access to its code or accounts. Same method for every product, and no AI decides the score. Last checked 2026-10-11.
Put together from public information, without Solheim's involvement. If anything here is wrong, tell us and a person will check it.
Visit Solheim → · Alternatives to Solheim → · How this was measured →