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.

Gemini 3.7 Flash vs Gemini 3.8 Flash for Maths and reasoning on a budget, as of 2026-09-23

Find an active text model for maths and reasoning while keeping one task at or under $0.08.

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

What the evidence says

The evidence can't separate them for this job.

Gemini 3.8 Flash has a 32% chance of matching or beating Gemini 3.7 Flash on this job's objective.

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

Decision status: partial. P(best) and score intervals retain the full ranked lineup for this job, including models beyond this pair and models with thin evidence. Models with unknown objective values remain unranked and are outside this probability. P(best) is not a head-to-head probability.

Scores and score intervals use the objective's dimensionless normalised units. Cost per task is in USD for the selected offering and the applied spec's task size. When task size is omitted, the engine uses its documented default. Unavailable means the engine returned no value.

Model	Score	Score interval	P(best), ranked job lineup	Cost per task, USD	Selected offering: provider / region / tier	Capability estimates and intervals
Gemini 3.7 Flash	0.712712	[0.299602, 1.125822]	23%	0.045	google-gemini-api / global / standard	maths: -0.018394566; interval [-0.971933015, 0.935143883]; benchmarks 3; direct benchmarks 1
Gemini 3.8 Flash	0.522469	[0.219187, 0.82575]	5%	0.045	google-gemini-api / global / standard	maths: -0.457512924; interval [-1.157546025, 0.242520177]; benchmarks 5; direct benchmarks 2

At most $0.08 a task; maths still ranked.

Requirements

model.class = text-generator: The answer must be generated as text.
model.lifecycle = active: An active model is still supported by its lab or provider.
offering.cost_per_task <= 0.08: The offering must stay within the per-task budget.

Preferences

maths: Prefer stronger maths among offerings inside the budget.
-offering.cost_per_task: Prefer the cheaper task among offerings inside the budget.

Applied template spec

spec_version

where

model.class = text-generator
model.lifecycle = active
offering.cost_per_task <= 0.08

optimize

weights

maths: 0.5
-offering.cost_per_task: 0.5

Evidence for Gemini 3.7 Flash

Benchmark	Version	Measurement	Unit	Published interval	Evidence date	Measurement conditions	Provenance
arena_sc_industry_mathematical	text_style_control / industry_mathematical, latest split, revision 1880dbebff5b	1510.79	Arena score (Elo scale)	Unavailable	2026-09-13	Effort: Unavailable Subcategory: Unavailable Harness: Unavailable Measured by: independent_evaluator Sample size: Unavailable	Source
arena_sc_math	text_style_control / math, latest split, revision 1880dbebff5b	1521.51	Arena score (Elo scale)	Unavailable	2026-09-13	Effort: Unavailable Subcategory: Unavailable Harness: Unavailable Measured by: independent_evaluator Sample size: Unavailable	Source
frontiermath_tiers_1_3_v2	FrontierMath-Tiers-1-3-v2-Private (Epoch AI run)	71.58	percent	Unavailable	2026-08-14	Effort: high Subcategory: Unavailable Harness: Unavailable Measured by: independent_evaluator Sample size: Unavailable	Source

Missing facts for Gemini 3.7 Flash

model.hardware_fit_indeterminate: unknown
model.languages: unknown
model.parameters_total: unknown
origin.base_lineage: not_disclosed
origin.weights_hosting: not_disclosed
offering.api_model_id: unknown
offering.attestation.baa: not_disclosed
offering.attestation.soc2: not_disclosed
offering.private_deployment: unknown
offering.router_slug: unknown
offering.speed.throughput: unknown
offering.speed.time_to_first_token: unknown

Evidence for Gemini 3.8 Flash

Benchmark	Version	Measurement	Unit	Published interval	Evidence date	Measurement conditions	Provenance
arena_sc_industry_mathematical	text_style_control / industry_mathematical, latest split, revision 1880dbebff5b	1519.73	Arena score (Elo scale)	Unavailable	2026-09-13	Effort: Unavailable Subcategory: Unavailable Harness: Unavailable Measured by: independent_evaluator Sample size: Unavailable	Source
arxivmath	ArXivMath, MathArena Overall table	55.35	percent	Unavailable	2026-09-30	Effort: Unavailable Subcategory: Unavailable Harness: Unavailable Measured by: benchmark_author Sample size: Unavailable	Source
brokenarxiv	BrokenArXiv, MathArena Overall table	22.4	percent	Unavailable	2026-09-30	Effort: Unavailable Subcategory: Unavailable Harness: Unavailable Measured by: benchmark_author Sample size: Unavailable	Source
frontiermath_tiers_1_3_v2	FrontierMath-Tiers-1-3-v2-Private (Epoch AI run)	68.42	percent	Unavailable	2026-09-02	Effort: high Subcategory: Unavailable Harness: Unavailable Measured by: independent_evaluator Sample size: Unavailable	Source
hle	Humanity's Last Exam, Scale Labs leaderboard	44.52	percent	Unavailable	2026-09-30	Effort: Unavailable Subcategory: Unavailable Harness: Unavailable Measured by: independent_evaluator Sample size: Unavailable	Source

Missing facts for Gemini 3.8 Flash

model.hardware_fit_indeterminate: unknown
model.languages: unknown
model.parameters_total: unknown
origin.base_lineage: not_disclosed
origin.weights_hosting: not_disclosed
offering.api_model_id: unknown
offering.attestation.baa: not_disclosed
offering.attestation.soc2: not_disclosed
offering.private_deployment: unknown
offering.router_slug: unknown
offering.speed.throughput: unknown
offering.speed.time_to_first_token: unknown

What this template teaches

The budget is a Must; inside it, cost and maths share the ranking equally.

Decision dec_9bc6d57a565f250162eba662 on public snapshot snap_05a963d2477ac2fb.

Read the full dated job answer.

The live board uses current data. This link applies the template to its full lineup.

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 · Model comparisons · How we decide