Open-weight vs closed models
What the numbers actually say
Cost rankings built from our own rate table, case studies drawn from cited industry reporting, and comparisons that name where a closed frontier model still wins. Every figure below either comes from our own catalogue or is attributed to the publisher who measured it — never both at once, and never restated as our own finding when it isn’t.
We do not publish benchmark or quality leaderboards of our own — see what we do not do. Where a ranking below is ours, it ranks price, discount depth, context window or cache depth: things our own rate table actually measures.
- PricingCost optimizationOpen-weight
What a Fixed Monthly AI Budget Actually Buys in 2026
Three realistic budget tiers, worked by hand against this catalogue's own rate table, at one stated request shape -- how many requests and tokens $10, $50 and $250 a month actually buys on three callable models.
- ComparisonOpen-weightPricing
When a Closed Frontier Model Is Still the Right Call
Closed frontier models measurably lead reasoning-heavy benchmarks as of September 2026. Where that lead and a simpler operational model are worth the higher price -- and why our catalogue is not the answer for that reader.
- Cost optimizationComparisonPricing
Self-Hosting vs a Managed Open-Weight API: When Each Wins
Where the self-host breakeven actually sits, what self-hosting really costs once engineering time is priced in, and the honest cases where self-hosting wins -- this is not a blanket argument for a managed API.
- Cost optimizationPricingOpen-weight
The Real Cost of an AI Coding Agent: A Token Budget Breakdown
Why an agent's bill doesn't look like a chat bill -- a closed-frontier full-day usage pattern reported near $594/month, agent loops burning 5-30x an equivalent chat interaction, and one study's finding that 59.4% of an agent's tokens go to review, not writing.
- PricingCost optimizationOpen-weight
Context Window Economics: What a Bigger Context Window Actually Costs
What a bigger context window actually costs, worked by hand across three models and three document sizes -- the input rate applies to every token sent, whether or not the window could hold far more.
- Cost optimizationPricingOpen-weight
Prompt Caching: Why It's the Biggest Lever in Your LLM Bill
How prompt caching actually works, what independent research says it's worth (41-80% cost reduction in one study), and what it changes on this catalogue's own cached rates.
- Vendor lock-inOpen-weightCost optimization
The Hidden Cost of Vendor Lock-In: What Closed-API Deprecations Actually Cost You
What a forced closed-model deprecation actually costs on a naive migration -- and why the fact that open weights don't disappear is a real, if partial, answer to it.
- Case studyCost optimizationOpen-weight
Why Enterprises Are Switching to Open-Weight Models in 2026
Case studies and cost research behind the enterprise shift to open-weight models -- a documented 81% bill cut from one team's own routing change, a Deloitte-cited ~40% average, and the honest reason adoption still lags what the economics support.
- PricingMedia modelsOpen-weight
Video and Image Generation Models: What's Priced and What's Actually Callable
Every video and image model in this catalogue carries a real, published rate. Almost none of the video ones can be called today. Here is the honest gap between what's priced and what's servable, and why we publish both anyway.
- RankingsComparisonOpen-weight
The 2026 Open-Weight Landscape: What the Independent Reviews Actually Say
A roundup of what outside reviewers, pricing surveys and case studies actually report about open-weight models in 2026 — including where two of them disagree on the same number, and where closed frontier models still lead.
- LicensingOpen-weightPricing
Data Sovereignty and the Open-Weight Advantage
Enterprises are increasingly choosing open-weight models to keep proprietary data and deployment decisions in their own hands, not just to save money. What the sovereignty argument actually claims, and where our own catalogue's licences sit.
- ComparisonPricingOpen-weight
Kimi K3 vs GLM-5.2 vs DeepSeek V4: The Real 2026 Cost Comparison
Independent reviewers keep shortlisting the same three families for coding work. Here is what each actually costs at our rates, on a real coding-agent volume, and where price and capability stop agreeing.
- RankingsPricingOpen-weight
Open-Weight Model Pricing Ranked: The 2026 Discount Table
Every open-weight model we serve, ranked by how far its price sits below its publisher's own rate — on the plain rate card and on a real agent workload.