ModelSpec is an analysis of alternatives for AI models: it decides which models fit a job from your requirements, sourced benchmarks and real cost, and shows its work.
Coding agent on a budget
Find an active coding model with enough context while keeping one task at or under $0.05.
Worked example as of 2026-09-23, from the public data image.
At most $0.05 a task; coding still ranked.
Requirements
model.class = text-generator: Coding work needs a model that generates text and code.model.lifecycle = active: An active model is still supported by its lab or provider.model.context_window >= 200000: A large repository needs room for code, tests, and instructions.offering.cost_per_task <= 0.05: The offering must stay within the per-task budget.
Preferences
software_engineering: Prefer stronger evidence on real software-engineering tasks.-offering.cost_per_task: Among qualifying offerings, prefer the cheaper task.
Applied template spec
spec_version- 1
where- model.class = text-generator
- model.lifecycle = active
- model.context_window >= 200000
- offering.cost_per_task <= 0.05
optimizeweightssoftware_engineering- 0.6
-offering.cost_per_task- 0.4
Worked answer
Decision status: partial.
Tied on the evidence
google/gemini-3-7-flashgoogle/gemini-3-8-flash
probability bands: a model joins the leader when P(its score >= the leader's) >= 0.25 under the score posteriors; a weighted capability with an 80% interval wider than 2.8 is not enough evidence
Tie-breakers describe members of the tie. They do not establish a winner.
- Most independently measured tie-breaker: google/gemini-3-8-flash
This is a partial answer. Unknown facts or missing objective evidence can leave eligible candidates unranked.
probability bands: a model joins the leader when P(its score >= the leader's) >= 0.25 under the score posteriors; a weighted capability with an 80% interval wider than 2.8 is not enough evidence
Score and score interval use the objective's dimensionless normalised units. Cost per task is for the engine-selected offering and the applied spec's task size. When task size is omitted, the engine uses its documented default. P(best) is the engine probability. Unavailable means the engine did not return a value. Thin evidence is undetermined and is not a recommendation.
Best band
| Model | Score | Score interval | P(best) | Cost per task, USD | Capability estimates and intervals |
|---|---|---|---|---|---|
google/gemini-3-7-flash | 0.6 | [-1.248386, 2.448386] | 0.63671875 | 0.045 |
|
google/gemini-3-8-flash | 0.0 | [-1.549939, 1.549939] | 0.36328125 | 0.045 |
|
Unknown requirements
Some candidates may qualify. Missing facts:
model.context_windowoffering.cost_per_task
Engine warnings
- not_separable
Decision dec_e714b968d27e687c50bbf4db on public snapshot snap_05a963d2477ac2fb.
What this template teaches
Use Must for non-negotiable context and budget limits, then Prefer to trade coding strength against cost.
The live board uses current data. Its answer can change as evidence changes.
Human lookup is free on Decide. CLI, API, and MCP access requires a key. Access and pricing.