Workspace · step 2 of 5
Model Selection Workspace
Pick a use case to weight which verified fields matter, then narrow a source-backed shortlist by provider, lifecycle, context window, modality, pricing coverage, hosted availability, verification status, or freshness. This is not a recommendation engine — shortlist order is deterministic, never a ranking. New here? Start at /how-it-works.
Shortlist
Shortlist (2)
Shortlist order: verified field count → active lifecycle → source count → name. This is not a ranking; it is a deterministic order that surfaces well-sourced records first.
Create evidence brief for top shortlist → Briefs preserve evidence and data gaps; they do not choose for you.
| Model | Provider / creator | Lifecycle | Context | Output | Modalities | Pricing | Hosted | Sources | Gaps | Freshness | Suggested next action |
|---|---|---|---|---|---|---|---|---|---|---|---|
Llama 4 ScoutVerified | meta | active | 10,000,000 | — | text-in, image-in, text-out | — | Available | 1 | 2 | Fresh | Inspect the hosted-availability record and the third-party hosted pricing reference on the provider page. |
DeepSeek V4 ProVerified | deepseek | active | 1,000,000 | — | — | Verified | Available | 3 | 4 | Fresh | Review the verified pricing reference and re-check the freshness state before projecting cost. |
Decision checklist
- Is the model active (not deprecated or retired)?
- Is the context window verified against the vendor docs?
- Is the max-output limit verified, or null?
- Is first-party pricing verified, or only hosted available?
- If hosted: do you accept the billing platform's rate as a reference (not a live quote)?
- Is the provider's status surface observed?
- Are the source citations fresh, or do they appear on the reverification queue?
- Are the remaining data gaps acceptable for this workload?
Build date: 2026-05-24. Freshness states are computed deterministically against this date.