Skip to content

Cheapest Decision Models per 1,000 Decisions

Cost per 1,000 decisions is the hosted input price at 200 input tokens per decision, the size of a typical state plus questions. Models below the median Decision Index score are left out, so the list is the cheapest of the competent ones; hosted models the Index has not scored yet are kept.

Ranked list (top 25)

#ModelDecision IndexECEp50 latencyPer 1k decisionsParams
1Baseten: Elephant 1 (free)
Baseten
———Free—
2Baseten: Mammoth 1 (free)
Baseten
———Free—
3Mercury Decide
Inception
———Free—
4SimpleJev Qwen3.6-35B-A3B (Featherless)
Featherless AIOpen weights
———Free—
5Span-01 Lite
Respan
———Free—
6Perplexity: Decider Flash Latest
Perplexity
———$0.0020—
7Perplexity: Decider Flash V1.2
Perplexity
———$0.0020—
8Respan: Span-02
Respan
———$0.0020—
9Cloudflare clef-flash
CloudflareOpen weights
44.40.02553 ms$0.003810B
10Perplexity Decider v1.1 (27B)
PerplexityOpen weights
62.80.071104 ms$0.004028B
11Liquid d1 (hosted)
Liquid AI
———$0.0040—
12Perplexity: Decider Latest
Perplexity
———$0.0040—
13Perplexity: Decider V1.2
Perplexity
———$0.0040—
14Span-01
Respan
———$0.0040—
15Kai 1
Hanzo AI
———$0.0042—
16meraGPT Decider 1
meraGPT (Okyasoft)
———$0.0060—
17Vansa-3.4
Vansa
———$0.0068—
18Nace.AI Drex v1.5
Nace.AIOpen weights
45.80.09054 ms$0.00809B
19Celeris-1 Decision
Celeris
———$0.0080—
20ScaleDown Classify-1
ScaleDown
———$0.0080—
21Jev
TypeSafe AI
60.20.080196 ms$0.0084—
22Cloudflare clef
CloudflareOpen weights
54.90.040102 ms$0.008428B
23LiquidAI: d1-ultra
LiquidAI
———$0.0084—
24Microsoft-Decision-1
Microsoft
———$0.0084—
25TypeSafe: Jev 1.14
TypeSafe
———$0.0084—

Scores, calibration (ECE) and latency from the Jev Decision Index; cost per 1,000 decisions at 200 input tokens each. Updated daily.

More decision-model rankings

Frequently asked questions

What is the cheapest decision model?

Baseten: Elephant 1 (free) is the cheapest hosted decision model in our data that is not below the median Decision Index score.

How is cost per 1,000 decisions calculated?

Input price per 1M tokens × 200 tokens × 1,000 decisions. Decision models return probabilities rather than text, so output tokens are negligible.

Is self-hosting an open decision model cheaper?

At high volume, usually: small open models run on a CPU, and the cost is the compute you already pay for. See the open-weights and CPU rankings.