Estimate. Plan. Win.

Open source. Honest pricing.

Bidwright is fully open source. Run it on your own machine for free. Or let us host it — same code, same features, plus the ops we'd run anyway. Bring your own LLM key, or use ours.

BYOK · pay your providerPooled AIComing soon

Self-host

Run the full platform on your own infrastructure. Same code, same features. AGPL-3.0.

Free/ forever
Read the docs
  • All product features — no feature gating
  • Multi-tenant orgs & users
  • Bring your own LLM key (any provider)
  • MCP server included
  • Full Postgres + pgvector access
  • Community support · GitHub Issues
Most popular

Hosted Cloud

We run Bidwright for you. Backups, encryption, updates, monitoring — your team just estimates.

$79/ per seat / month

5 seats minimum · annual or monthly billing

Join early access
  • Everything in Self-host, fully managed
  • Daily encrypted backups · point-in-time restore
  • Zero-downtime updates
  • Email + branded PDF delivery included
  • Bring your own LLM key (any provider)
  • Business-hours support · 99.9% uptime SLO

Enterprise

For estimating teams that need SSO, dedicated infrastructure, custom data residency, and an SLA you can sign.

Custom/ annual
Talk to us
  • Everything in Hosted Cloud
  • SSO (SAML, OIDC) · SCIM provisioning
  • Dedicated database & compute
  • Data residency: US / Canada / EU
  • 24×7 priority support · 99.95% SLA
  • Custom plugins, agents, integrations
  • Onboarding & migration services

Bring your own key

No middleman markup on AI usage.

Drop your Anthropic, OpenAI, OpenRouter, or Gemini key into the org settings. Bidwright uses it for the agent, takeoff Ask-AI, knowledge ingestion, and quote review. Your provider bills you directly. We never proxy.

Want to run fully offline? Use LM Studio + Ollama embeddings. No outbound calls.

Approximate AI cost per estimate

Mid-sized commercial bid. Token use varies by document volume.

Provider / modelPer estimate
Claude Sonnet 4.6 (cached)$0.40 – $1.20
Claude Opus 4.7 (review only)$1.50 – $4.00
GPT-5 (mixed)$0.60 – $1.80
Gemini 2.5 Pro$0.30 – $0.90
Local (Qwen 72B + Ollama)$0 marginal

Prompt caching enabled by default. Most users see 60–80% cache hit rates on knowledge searches.

Add-ons (Hosted)

Extra storage
$0.10 / GB / mo

Bid packages, drawings, models, exports. First 25 GB / seat included.

Custom branded PDFs
$25 / one-time

Logo, color, layout — your quote, your face.

Onboarding session
$500 / 90 min

Migrate templates, rate schedules, and assemblies with us.

Pricing FAQ

Do I need to pay for AI usage on top of my seat?

Yes — but you pay your provider directly, not us. Bidwright is BYOK: drop your Anthropic, OpenAI, OpenRouter, or Gemini key into the org settings and your provider invoices you for usage. A pooled AI add-on (so you don't have to manage keys) is on the roadmap but not available today.

Is the open-source version really the same?

Yes. Same code, same features. The hosted plan is the operations work — backups, scaling, monitoring, updates, support. We do not feature-gate the open-source build.

Can I migrate from hosted back to self-host?

Always. Export your tenant any time — Postgres dump + object storage bundle. The same migration tools work in both directions.

What about AGPL and SaaS use?

If you embed Bidwright in your own SaaS, AGPL applies — you must offer the source to your users. Talk to us about a commercial license if AGPL doesn't fit.

Do you offer a discount for annual?

Yes — 2 months free on annual hosted billing. Enterprise pricing is annual only.

What does the cost include for compute?

Hosted seats include shared compute, Postgres, pgvector, Redis, and 25 GB of storage per seat. We scale it for you.