Instamodel
Manage keysTry public demo
Skip to model results
Unified model catalog

Every model.
One route.

Compare frontier APIs, deployable open weights, and models post-trained for the job. Try the deterministic public demo without a key; paid routes remain release-gated.

Models listed
13
Specialists
5
Inference partners
17
API schema
OpenAI-compatible

Catalog reviewed August 17, 2026. Pricing, latency, provider routes, and specialist evaluations are illustrative prototype data.

01 / FIND A MODEL

Route with evidence.

Specialist publishers in this prototype are intentionally marked as fictional. Foundation family names identify their real-world publishers; marketplace economics remain illustrative.

Model type
Showing 13 of 13 modelsIllustrative rates · paid routes not purchasable yet
01Frontier
Anthropic

Claude Fable 5

Foundation publisher

Anthropic's highest-capability widely released model for long-running agents and enterprise work.

GeneralCodingAgenticTextVision
Context
1M
Input
$10.00
Output
$50.00
Median
1860ms
Network qualityFrontier
Publisher model; no relay-authored score
Served by

Anthropic · Amazon Bedrock · Google Cloud

Compare
02Post-trained
Northstar Legal

ClauseGuard 32B

Fictional demo publisher

Contract-review specialist tuned to flag policy deviations and cite source clauses.

LegalExtractionText
Post-trained fromMistral Small 4
Context
128K
Input
$0.32
Output
$0.78
Median
330ms
CUAD F191.4
Fictional publisher-submitted demo result
Served by

Baseten · Fireworks AI

Compare
03Frontier
Google

Gemini 3.6 Flash

Foundation publisher

Fast multimodal intelligence tuned for high-throughput agents and document workloads.

GeneralMultimodalAgenticTextVisionAudioVideo
Context
1M
Input
$1.50
Output
$7.50
Median
540ms
Network qualityFrontier
Publisher model; no relay-authored score
Served by

Google AI · Vertex AI

Compare
04Frontier
OpenAI

GPT-5.6 Sol

Foundation publisher

General-purpose reasoning for complex agentic, coding, and knowledge workflows.

GeneralCodingAgenticTextVision
Context
1.05M
Input
$5.00
Output
$30.00
Median
1420ms
Network qualityFrontier
Publisher model; no relay-authored score
Served by

OpenAI

Compare
05Open weights
Meta

Llama 4 Maverick

Foundation publisher

Natively multimodal mixture-of-experts model with strong quality-to-cost routing.

GeneralMultimodalCodingTextVision
Context
1M
Input
$0.22
Output
$0.88
Median
410ms
Network availability3 routes
Illustrative provider availability
Served by

Fireworks AI · Together AI · Nebius

Compare
06Post-trained
Lattice Commerce

CatalogMatch 8B

Fictional demo publisher

Commerce specialist for product normalization, taxonomy mapping, and duplicate detection.

CommerceClassificationTextVision
Post-trained fromLlama 4 Scout
Context
64K
Input
$0.08
Output
$0.24
Median
180ms
Catalog F197.1
Fictional publisher-submitted demo result
Served by

Groq · Together AI

Compare
07Open weights
OpenAI

gpt-oss-120b

Foundation publisher

Large open-weight reasoning model for customizable, self-hostable workloads.

GeneralCodingAgenticText
Context
131K
Input
$0.10
Output
$0.50
Median
350ms
Network availability3 routes
Illustrative provider availability
Served by

Groq · Fireworks AI · Together AI

Compare
08Frontier
xAI

Grok 4.5

Foundation publisher

Frontier reasoning for coding, tool use, and knowledge-intensive agent workflows.

GeneralCodingAgenticTextVision
Context
500K
Input
$2.00
Output
$6.00
Median
980ms
Network qualityFrontier
Publisher model; no relay-authored score
Served by

xAI

Compare
09Post-trained
Instamodel

Instamodel Demo

Deterministic platform demo

Deterministic sandbox route for testing authentication, request receipts, and the OpenAI-compatible response shape without contacting an inference provider.

SandboxIntegrationText
Context
16K
Input
$0.00
Output
$0.00
Median
8ms
ExecutionDemo only
Not a model-quality result
Served by

Local deterministic demo

Compare
10Post-trained
Shelfwise Studio

MerchCopy 24B

Fictional demo publisher

Commerce specialist for product-attribute enrichment, merchandising copy, and taxonomy cleanup.

CommerceContentTextVision
Post-trained fromMistral Small 4
Context
192K
Input
$0.29
Output
$0.74
Median
290ms
Attribute F194.2
Fictional publisher-submitted demo result
Served by

Baseten · Runpod

Compare
11Open weights
Mistral AI

Mistral Large 3

Foundation publisher

Open-weight general-purpose multimodal model for private and regional deployment.

GeneralMultimodalCodingTextVision
Context
256K
Input
$0.50
Output
$1.50
Median
520ms
Network availability3 routes
Illustrative provider availability
Served by

Mistral Compute · Together AI · Fireworks AI

Compare
12Open weights
Mistral AI

Mistral Medium 3.5

Foundation publisher

Frontier-class multimodal model optimized for agentic and coding use cases.

GeneralCodingMultimodalTextVision
Context
256K
Input
$1.50
Output
$7.50
Median
690ms
Network qualityFrontier
Publisher model; no relay-authored score
Served by

Mistral Compute · Azure AI

Compare
13Post-trained
HelioHelp Labs

SupportResolve 14B

Fictional demo publisher

Customer-support specialist for grounded answer drafts, intent labels, and escalation summaries.

Customer supportExtractionText
Post-trained fromQwen3 14B
Context
128K
Input
$0.38
Output
$0.92
Median
370ms
Resolve rate86.7
Fictional publisher-submitted demo result
Served by

Modal · Nebius

Compare