Proprietary AI models: API-only, and what that locks you into

Prices from official vendor price lists · October 2026
151 closed-weight models available only through the vendor API — no self-hosting, no infrastructure to run. Output prices run from $0.063 to $180 per 1M tokens across the 117 models with a paid per-token price; 23 have a free tier.
Proprietary API
Services found: 151
#AI modelCapabilitiesAPI $/1M (in / out)Context windowPrice / moOutput $/1MOverviewVisit site
41

Mercury 2

Inception
Text & chatMultilingualProprietary API
$0.25 / $0.75128K tokensAPI only0.75View
42

Trinity Large Thinking

Arcee Ai
Text & chatReasoningTool calling
$0.25 / $0.8262.1K tokensAPI only0.8View
43

Mistral Large 3 2512

Mistral AI
Text & chatVisionReasoning
$0.5 / $1.5262.1K tokensAPI only1.5View
44

Devstral 2 2512

Mistral AI
Text & chatCodingStructured JSON output
$0.4 / $2262.1K tokensAPI only2View
45

GPT-5.4

OpenAI
Text & chatVisionReasoning
$2.5 / $151.05M tokensAPI only15View
46

GPT-5.3-Codex

OpenAI
Text & chatVisionCoding
$1.75 / $14400K tokensAPI only14View
47

GPT-Realtime-2.1

OpenAI
Text & chatVisionAudio
$4 / $24128K tokensAPI only24View
48

GPT-Realtime-2.1 Mini

OpenAI
Text & chatVisionAudio
$0.6 / $2.4128K tokensAPI only2.4View
49

Gemini 3.1 Flash-Lite

Google
Text & chatVisionAudio
$0.25 / $1.51.05M tokensAPI only1.5View
50

Nano Banana 2

Google
VisionImage generationStructured JSON output
API onlyView
51

Nano Banana 2 Lite

Google
VisionImage generationStructured JSON output
API onlyView
52

Nano Banana Pro

Google
VisionImage generationStructured JSON output
$2 / $12API only12View
53

Gemini Omni Flash

Google
Text & chatVisionAudio
$1.5 / $9API only9View
54

Gemini 3.5 Transcribe

Google
AudioAdjustable thinkingSpeech-to-text
API onlyView
55

Gemini 3.5 Transcribe Live

Google
AudioAdjustable thinkingSpeech-to-text
API onlyView
56

Grok 4.20 Reasoning

xAI
Text & chatVisionReasoning
$1.25 / $2.51M tokensAPI only2.5View
57

Grok 4.20 Non-Reasoning

xAI
Text & chatVisionReasoning
$1.25 / $2.51M tokensAPI only2.5View
58

Grok Build 0.1

xAI
Text & chatAdjustable thinkingWeb search grounding
$1 / $2256K tokensAPI only2View
59

Codestral

Mistral AI
Text & chatCodingStructured JSON output
$0.3 / $0.9128K tokensAPI only0.9View
60

Devstral 2

Mistral AI
Text & chatCodingStructured JSON output
256K tokensAPI onlyView

Browse by need

Reasoning models169 modelsCoding assistants17 modelsVision and images156 modelsLong context 1M+82 modelsFree tier available80 modelsOpen weights88 models

These 151 models are available only through the vendor own API: no weights to download, no self-hosting, no fine-tuning beyond what the vendor exposes. In exchange you get no infrastructure to run and immediate access to whatever ships next.

Prices run from $0.87 to $50 per million output tokens. The frontier tier is entirely proprietary - GPT-6 Astra, Claude Fable 5.1, Claude Opus 5 - and that gap is the honest argument for closed models. The arguments against are equally practical: pricing changes without your consent, models get retired on the vendor schedule, and your data leaves your perimeter.

No infrastructureNothing to provision, patch or scale — the vendor absorbs all of it.
The frontier lives hereEvery top-quality model today is closed. There is no open equivalent yet.
Vendor risk is realPrices change, models retire, and your data crosses a boundary.

Models with this capability

The 151 models below all support this capability. Prices are input and output per 1M tokens from the vendor official price list; for dirhams multiply by 3.6725. Tap a card to open the model page.

Claude Fable 5.1
#1
Anthropic
in $10out $501M tokens
GPT-6 Astra
#2
OpenAI
in $10out $501.05M tokens
GPT-6.1 Sol
#3
OpenAI
in $2out $101.05M tokens
GPT-6 Sol
#4
OpenAI
in $2out $101.05M tokens
GPT-6 Luna
#5
OpenAI
in $0.1out $0.51.05M tokens
Claude Opus 5.5
#6
Anthropic
in $4out $201M tokens
Claude Sonnet 5.5
#7
Anthropic
in $2out $101M tokens
Grok 4.7
#8
xAI
in $2out $6500K tokens

How to choose

  • Check the deprecation policy before building on a specific model version.
  • Pin snapshot versions where the vendor offers them, rather than floating aliases.
  • Read the data-retention terms if you handle personal or regulated information.
  • Keep an open-weight fallback identified, even if you never use it.

Common mistakes

Building on a floating aliasA silent model update can change behaviour overnight.
No exit planTeams discover lock-in at the moment prices rise, not before.

Who this suits

Small teamsNo GPU budget, no ops capacity, immediate access to the best models.
EnterprisesCompliance certifications and support contracts that self-hosting cannot match.
PrototypingValidating an idea before committing infrastructure.

Related filters

Open weights88 models you can download and self-host
Frontier tier41 flagship models, the best each vendor sells
Priority tier65 models with a paid fast lane
Prompt caching147 models with cheap repeated context
All 239 modelsThe full rating with filters and prices

Frequently asked questions

Why choose a closed model?

The frontier tier is entirely proprietary. If you need the best available quality, there is no open alternative yet.

What are the risks?

Pricing changes without your consent, models get retired on the vendor schedule, and your data leaves your perimeter.

What is the open alternative?

90 models in this rating publish weights — GLM, Qwen, DeepSeek, Mistral, Gemma and others.

Can I fine-tune a proprietary model?

Sometimes, within the vendor platform. You never receive the weights.

What happens when a model retires?

You migrate. Vendors publish timelines, but the work lands on you.

Frequently searched

By capabilityReasoningCodingVisionLong context 1M+Free tierOpen weightsLow latencyRealtime voiceImage generation
By vendorAnthropic (13)OpenAI (30)Google DeepMind (17)xAI (8)Meta (6)DeepSeek (9)Mistral AI (11)Alibaba (Qwen) (34)Z.ai (12)Moonshot AI (5)
Popular modelsClaude Fable 5.1GPT-6 AstraClaude Opus 5Claude Fable 5GPT-5.6 SolGLM 5.3Grok 4.6Kimi K3GPT-5.6 Terra
By deliveryBatch APIPrompt cachingPriority tierFrontier tierProprietary APITool callingStructured JSONWeb searchMultilingual

Ratings 0

Avg. score 0

Rate this page

Help us improve