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.

Maths and reasoning, self-hosted

Find a commercially usable open-weights model for maths and reasoning that you can run yourself.

Worked example as of 2026-09-23, from the public data image.

Open weights you run yourself; commercial licence.

Requirements

model.class = text-generator: The answer must be generated as text.
model.weights_openness = open_weights: Self-hosting requires downloadable weights; add a model.fits_hardware Must for the device you own.
licence.commercial_use in {permitted, permitted_with_conditions}: The licence must allow commercial use, including use with stated conditions.

Preferences

maths: Prefer stronger maths among self-hostable models.
reasoning: Prefer stronger multi-step reasoning among self-hostable models.

Applied template spec

spec_version

where

model.class = text-generator
model.weights_openness = open_weights
licence.commercial_use in {permitted, permitted_with_conditions}

optimize

weights

maths: 0.6
reasoning: 0.4

Worked answer

Decision status: partial.

No recommendation: the public image has no answer with enough evidence.

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.

Thin evidence: undetermined

Model	Score	Score interval	P(best)	Cost per task, USD	Capability estimates and intervals
`moonshot/kimi-k3`	0.820499	[-0.335143, 1.976142]	0.359375	Unavailable	maths: -0.032135085; interval [-0.762881535, 0.698611364]; benchmarks 4; direct benchmarks 2 reasoning: 0.00591179; interval [-1.716605348, 1.728428927]; benchmarks 2; direct benchmarks 0
`zhipu/glm-5-3`	0.701608	[-0.570189, 1.973406]	0.296875	0.0736	maths: -0.156897477; interval [-1.103624192, 0.789829238]; benchmarks 3; direct benchmarks 1 reasoning: -0.105195063; interval [-1.810826758, 1.600436632]; benchmarks 2; direct benchmarks 0
`moonshot/kimi-k2-6`	0.318442	[-0.855888, 1.492771]	0.12109375	Unavailable	maths: -0.556964019; interval [-1.297343038, 0.183415001]; benchmarks 4; direct benchmarks 2 reasoning: -0.465971358; interval [-2.219253326, 1.28731061]; benchmarks 2; direct benchmarks 0
`deepseek/deepseek-v4-pro`	0.197978	[-0.967487, 1.363444]	0.1171875	0.06864	maths: -0.735823404; interval [-1.474402989, 0.00275618]; benchmarks 4; direct benchmarks 2 reasoning: -0.508628162; interval [-2.24362481, 1.226368486]; benchmarks 2; direct benchmarks 0
`zhipu/glm-5-2`	0.10002	[-1.063743, 1.263783]	0.10546875	0.0736	maths: -1.082359497; interval [-1.820422979, -0.344296015]; benchmarks 4; direct benchmarks 2 reasoning: -0.275228659; interval [-2.006940496, 1.456483179]; benchmarks 2; direct benchmarks 0

Unknown requirements

Some candidates may qualify. Missing facts:

maths
model.weights_openness

Engine warnings

not_separable

Decision dec_220085cc06f746f54108d436 on public snapshot snap_05a963d2477ac2fb.

What this template teaches

Privacy is a Must on the weights and the licence; the canvas shows the model size you would have to host. Lab jurisdiction is a buyer-set Must (origin.lab_jurisdiction in {…}); an unverified jurisdiction does not pass, so set the countries on the facet board. The same condition works as an inventory profile rule.

The live board uses current data. Its answer can change as evidence changes.

Apply this template on Decide

Human lookup is free on Decide. CLI, API, and MCP access requires a key. Access and pricing.

Use-case examples · How we decide