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.

General assistant, best available

Find the active assistant whose open-ended responses people rate most highly.

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

Most preferred responses, cost no object.

Requirements

model.class = text-generator: A conversational assistant needs a text-generating model.
model.lifecycle = active: An active model is still supported by its lab or provider.

Preferences

chat_preference: Rank only by the chat-preference capability estimate.

Applied template spec

spec_version

where

model.class = text-generator
model.lifecycle = active

optimize

weights

chat_preference: 1.0

Worked answer

Decision status: partial.

Separated on the evidence

anthropic/claude-opus-5-5

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

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
`anthropic/claude-opus-5-5`	1.0	[0.760212, 1.239788]	0.63671875	0.24	chat_preference: 1.584232494; interval [0.565758585, 2.602706403]; benchmarks 1; direct benchmarks 1

Rest band

Model	Score	Score interval	P(best)	Cost per task, USD	Capability estimates and intervals
`anthropic/claude-fable-5`	0.870848	[0.835874, 0.905823]	0.1015625	0.6	chat_preference: 1.035674964; interval [0.887124662, 1.184225266]; benchmarks 34; direct benchmarks 34
`anthropic/claude-opus-4-6`	0.80955	[0.775557, 0.843543]	0.01171875	0.3	chat_preference: 0.775317472; interval [0.630936802, 0.919698141]; benchmarks 36; direct benchmarks 36
`anthropic/claude-fable-5-1`	0.809424	[0.772803, 0.846044]	0.00390625	0.6	chat_preference: 0.774780684; interval [0.619237933, 0.930323435]; benchmarks 31; direct benchmarks 31
`anthropic/claude-opus-4-7`	0.784135	[0.749161, 0.81911]	0.0	0.3	chat_preference: 0.667370368; interval [0.518820066, 0.81592067]; benchmarks 34; direct benchmarks 34
`openai/gpt-6-sol`	0.768774	[0.528986, 1.008562]	0.07421875	0.12	chat_preference: 0.602126414; interval [-0.416347496, 1.620600323]; benchmarks 1; direct benchmarks 1
`qwen/qwen3-8-max-0902`	0.766008	[0.52622, 1.005796]	0.078125	0.104	chat_preference: 0.59037787; interval [-0.428096039, 1.60885178]; benchmarks 1; direct benchmarks 1
`anthropic/claude-opus-5`	0.750015	[0.715041, 0.78499]	0.0	0.3	chat_preference: 0.522450073; interval [0.373899771, 0.671000375]; benchmarks 34; direct benchmarks 34
`google/gemini-3-8-flash`	0.736496	[0.694902, 0.77809]	0.0	0.045	chat_preference: 0.465029132; interval [0.288363151, 0.641695113]; benchmarks 24; direct benchmarks 24
`google/gemini-3-7-flash`	0.721666	[0.680072, 0.76326]	0.0	0.045	chat_preference: 0.402038706; interval [0.225372725, 0.578704687]; benchmarks 24; direct benchmarks 24
`moonshot/kimi-k3`	0.695988	[0.658131, 0.733845]	0.0	Unavailable	chat_preference: 0.292976186; interval [0.132183141, 0.453769231]; benchmarks 29; direct benchmarks 29
`qwen/qwen3-8-flash-next`	0.682171	[0.442383, 0.92196]	0.04296875	Unavailable	chat_preference: 0.234290592; interval [-0.784183317, 1.252764502]; benchmarks 1; direct benchmarks 1
`openai/gpt-5-6-sol`	0.678586	[0.643077, 0.714094]	0.0	0.24	chat_preference: 0.219061677; interval [0.068242727, 0.369880626]; benchmarks 33; direct benchmarks 33
`xai/grok-4-7`	0.675244	[0.435455, 0.915032]	0.05078125	0.104	chat_preference: 0.204866691; interval [-0.813607219, 1.2233406]; benchmarks 1; direct benchmarks 1
`meta/muse-spark-1-3`	0.665866	[0.611551, 0.720181]	0.0	0.067	chat_preference: 0.165034925; interval [-0.065661987, 0.395731837]; benchmarks 13; direct benchmarks 13
`zhipu/glm-5-3`	0.663134	[0.625277, 0.700991]	0.0	0.0736	chat_preference: 0.153431285; interval [-0.007361761, 0.31422433]; benchmarks 29; direct benchmarks 29
`openai/gpt-6-astra`	0.636438	[0.596467, 0.676409]	0.0	0.6	chat_preference: 0.040043976; interval [-0.12972742, 0.209815373]; benchmarks 26; direct benchmarks 26
`google/gemini-3-1-pro-preview`	0.629612	[0.594637, 0.664586]	0.0	0.128	chat_preference: 0.01105026; interval [-0.137500042, 0.159600562]; benchmarks 34; direct benchmarks 34
`openai/gpt-5-4`	0.589281	[0.555288, 0.623274]	0.0	0.16	chat_preference: -0.16025079; interval [-0.304631459, -0.01587012]; benchmarks 36; direct benchmarks 36
`deepseek/deepseek-v3-1`	0.572739	[0.533498, 0.611981]	0.0	Unavailable	chat_preference: -0.230508781; interval [-0.397181294, -0.063836267]; benchmarks 27; direct benchmarks 27
`google/gemini-3-5-flash`	0.563107	[0.528132, 0.598081]	0.0	0.096	chat_preference: -0.271422803; interval [-0.419973105, -0.122872501]; benchmarks 34; direct benchmarks 34
`meta/muse-spark-1-1`	0.557072	[0.510359, 0.603786]	0.0	0.067	chat_preference: -0.297052895; interval [-0.4954636, -0.098642189]; benchmarks 19; direct benchmarks 19
`zhipu/glm-5-2`	0.529947	[0.49209, 0.567804]	0.0	0.0736	chat_preference: -0.412265302; interval [-0.573058347, -0.251472257]; benchmarks 29; direct benchmarks 29
`google/gemini-2-5-flash`	0.491228	[0.453993, 0.528463]	0.0	0.022	chat_preference: -0.576719849; interval [-0.734872721, -0.418566977]; benchmarks 34; direct benchmarks 34
`deepseek/deepseek-v4-pro`	0.440159	[0.402302, 0.478016]	0.0	0.06864	chat_preference: -0.793630284; interval [-0.954423329, -0.632837239]; benchmarks 29; direct benchmarks 29
`moonshot/kimi-k2-6`	0.431412	[0.396438, 0.466387]	0.0	Unavailable	chat_preference: -0.830780407; interval [-0.97933071, -0.682230105]; benchmarks 34; direct benchmarks 34
`microsoft/phi-4`	0.0	[-0.039985, 0.039985]	0.0	Unavailable	chat_preference: -2.663154702; interval [-2.832988405, -2.493320999]; benchmarks 26; direct benchmarks 26

Unknown requirements

Some candidates may qualify. Missing facts:

chat_preference

Engine warnings

not_separable

Omitted result rows: models 17; offerings 18. The bands above include all ranked models.

Decision dec_7f2a6357d164309eef5edbee on public snapshot snap_05a963d2477ac2fb.

What this template teaches

One Must selects the class, one Prefer states what best means.

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