Supported Models - concentrate

Overview

Concentrate provides access to 130+ models from 19 authors across 16 providers through a single unified API. Use any model slug below in the model field of your request.

{
  "model": "claude-opus-4-7",
  "input": "Hello, world!"
}

You can also pin a specific provider using the provider/model format:

{
  "model": "anthropic/claude-opus-4-7",
  "input": "Hello, world!"
}

Pricing, context windows, and provider availability are kept current in the Model Fortress dashboard. You can also query the List Models endpoint for real-time data.

Models by Author

OpenAI (23 models)

Frontier

Model Slug	Name	Providers	Context	Max Output
`gpt-5.5`	GPT 5.5	openai, azure	1,050,000	128,000
`gpt-5.4`	GPT 5.4	openai, azure	1,050,000	128,000
`gpt-5.4-pro`	GPT 5.4 Pro	openai, azure	1,050,000	128,000
`gpt-5.4-mini`	GPT 5.4 Mini	openai, azure	128,000	128,000
`gpt-5.4-nano`	GPT 5.4 Nano	openai, azure	128,000	128,000
`gpt-5.2`	GPT 5.2	openai, azure	400,000	128,000
`gpt-5.1`	GPT 5.1	openai, azure	400,000	128,000
`gpt-5`	GPT 5	openai, azure	400,000	128,000
`gpt-5-mini`	GPT 5 Mini	openai, azure	400,000	128,000
`gpt-5-nano`	GPT 5 Nano	openai, azure	400,000	128,000

Codex (Agentic Coding)

Model Slug	Name	Providers	Context	Max Output
`gpt-5.3-codex`	GPT 5.3 Codex	openai, azure	400,000	128,000
`gpt-5.2-codex`	GPT 5.2 Codex	openai	400,000	128,000
`gpt-5.1-codex-max`	GPT 5.1 Codex Max	openai	400,000	128,000
`gpt-5.1-codex-mini`	GPT 5.1 Codex Mini	openai	400,000	128,000

Reasoning

Model Slug	Name	Providers	Context	Max Output
`o1`	OpenAI o1	openai	200,000	100,000

Previous Generation

Model Slug	Name	Providers	Context	Max Output
`gpt-4.1`	GPT 4.1	openai, azure	1,047,576	32,768
`gpt-4.1-mini`	GPT 4.1 Mini	openai, azure	1,047,576	32,768
`gpt-4o`	GPT 4o	openai, azure	128,000	16,384
`gpt-4o-mini`	GPT 4o Mini	openai, azure	128,000	16,384

Open-Weight (GPT-OSS)

Model Slug	Name	Providers	Context	Max Output
`gpt-oss-120b`	OpenAI gpt-oss 120B	azure, cloudflare, bedrock, bluelobster	131,072	128,000
`gpt-oss-20b`	OpenAI gpt-oss 20B	cloudflare, bedrock, bluelobster	131,072	128,000
`gpt-oss-safeguard-120b`	OpenAI gpt-oss Safeguard 120B	bedrock	128,000	128,000
`gpt-oss-safeguard-20b`	OpenAI gpt-oss Safeguard 20B	bedrock	128,000	128,000

Anthropic (10 models)

Anthropic models support Zero Data Retention when accessed via the Anthropic provider, except where noted in the ZDR column below.

Current Generation

Model Slug	Name	Providers	Context	Max Output	ZDR
`claude-fable-5`	Claude Fable 5	anthropic	1,000,000	128,000	No
`claude-opus-4-7`	Claude Opus 4.7	anthropic, azure, bedrock	1,000,000	128,000	Yes
`claude-opus-4-6`	Claude Opus 4.6	anthropic, vertex, azure, bedrock	200,000	128,000	Yes
`claude-sonnet-4-6`	Claude Sonnet 4.6	anthropic, vertex, azure, bedrock	200,000	64,000	Yes
`claude-opus-4-5`	Claude Opus 4.5	anthropic, azure, bedrock	200,000	64,000	Yes
`claude-sonnet-4-5`	Claude Sonnet 4.5	anthropic, azure, bedrock	200,000	64,000	Yes
`claude-haiku-4-5`	Claude Haiku 4.5	anthropic, azure, bedrock	200,000	64,000	Yes

Previous Generation

Model Slug	Name	Providers	Context	Max Output	ZDR
`claude-opus-4-1`	Claude Opus 4.1	anthropic, bedrock	200,000	32,000	Yes
`claude-sonnet-4`	Claude Sonnet 4	anthropic, bedrock	200,000	64,000	Yes
`claude-opus-4`	Claude Opus 4	anthropic	200,000	32,000	Yes

Google (10 models)

Gemini

Model Slug	Name	Providers	Context	Max Output
`gemini-3.1-pro-preview`	Gemini 3.1 Pro Preview	vertex	1,000,000	65,536
`gemini-3.1-flash-lite-preview`	Gemini 3.1 Flash Lite Preview	vertex, ai-studio	1,048,576	65,536
`gemini-3-flash-preview`	Gemini 3 Flash Preview	vertex, ai-studio	1,000,000	65,536
`gemini-2.5-pro`	Gemini 2.5 Pro	vertex	1,000,000	65,536
`gemini-2.5-flash`	Gemini 2.5 Flash	vertex	1,000,000	65,536

Gemma (Open-Weight)

Model Slug	Name	Providers	Context	Max Output
`gemma-4-31b`	Gemma 4 31B	deepinfra	262,144	16,384
`gemma-4-26b-a4b`	Gemma 4 26B A4B	deepinfra	262,144	16,384
`gemma-3-27b`	Gemma 3 27B	bedrock	128,000	8,192
`gemma-3-12b`	Gemma 3 12B	cloudflare, bedrock	128,000	8,192
`gemma-3-4b`	Gemma 3 4B IT	bedrock	128,000	8,192

xAI (13 models)

Grok 4.20

Model Slug	Name	Providers	Context	Max Output
`grok-4.20-multi-agent-0309`	Grok 4.20 Multi-Agent	xai	2,000,000	131,072
`grok-4.20-0309-reasoning`	Grok 4.20 Reasoning	xai	2,000,000	131,072
`grok-4.20-0309-non-reasoning`	Grok 4.20 Non Reasoning	xai	2,000,000	131,072

Grok 4 & 4.x

Model Slug	Name	Providers	Context	Max Output
`grok-4.3`	Grok 4.3	xai	1,000,000	1,000,000
`grok-4-1-fast-reasoning`	Grok 4.1 Fast Reasoning	xai	2,000,000	131,072
`grok-4-1-fast-non-reasoning`	Grok 4.1 Fast	xai	2,000,000	131,072
`grok-4`	Grok 4	xai	256,000	256,000
`grok-4-0709`	Grok 4 (0709)	xai	256,000	131,072
`grok-4-fast-reasoning`	Grok 4 Fast Reasoning	xai	2,000,000	131,072
`grok-4-fast-non-reasoning`	Grok 4 Fast	xai	2,000,000	131,072

Grok 3

Model Slug	Name	Providers	Context	Max Output
`grok-3`	Grok 3	xai	131,072	131,072
`grok-3-mini`	Grok 3 Mini	xai	131,072	131,072

Code

Model Slug	Name	Providers	Context	Max Output
`grok-code-fast-1`	Grok Code Fast	xai	256,000	131,072

Meta (12 models)

Llama 4

Model Slug	Name	Providers	Context	Max Output
`llama-4-maverick`	Llama 4 Maverick	bedrock	1,048,576	8,192
`llama-4-scout`	Llama 4 Scout	cloudflare, bedrock	131,000	8,192
`llama-4-scout-17b-16e-instruct`	Llama 4 Scout 17B 16E Instruct	cloudflare	131,000	32,768

Llama 3.x

Model Slug	Name	Providers	Context	Max Output
`llama-3.3-70b-instruct`	Llama 3.3 70B Instruct	cloudflare, bedrock	128,000	8,192
`llama-3.2-90b-instruct`	Llama 3.2 90B Instruct	bedrock	128,000	8,192
`llama-3.2-11b-instruct`	Llama 3.2 11B Instruct	bedrock	128,000	8,192
`llama-3.2-3b-instruct`	Llama 3.2 3B Instruct	cloudflare, bedrock	128,000	8,192
`llama-3.2-1b-instruct`	Llama 3.2 1B Instruct	cloudflare, bedrock	128,000	8,192
`llama-3.1-70b-instruct`	Llama 3.1 70B Instruct	bedrock	128,000	8,192
`llama-3.1-8b-instruct`	Llama 3.1 8B Instruct	cloudflare, bedrock	128,000	8,192

Llama 3

Model Slug	Name	Providers	Context	Max Output
`llama-3-70b-instruct`	Llama 3 70B Instruct	bedrock	8,000	2,048
`llama-3-8b-instruct`	Llama 3 8B Instruct	bedrock	8,000	2,048

Mistral (13 models)

Frontier

Model Slug	Name	Providers	Context	Max Output
`mistral-large-3`	Mistral Large 3	mistral, bedrock	256,000	32,000
`mistral-medium-3.1`	Mistral Medium 3.1	mistral	128,000	32,000
`mistral-medium-3`	Mistral Medium 3	mistral	128,000	32,000

Magistral (Reasoning)

Model Slug	Name	Providers	Context	Max Output
`magistral-medium-1.2`	Magistral Medium 1.2	mistral	128,000	128,000
`magistral-small-1.2`	Magistral Small 1.2	mistral, bedrock	128,000	128,000

Code

Model Slug	Name	Providers	Context	Max Output
`codestral`	Codestral	mistral	128,000	32,000
`devstral-2`	Devstral 2	mistral	256,000	32,000

Small & Efficient

Model Slug	Name	Providers	Context	Max Output
`mistral-small-3.2`	Mistral Small	mistral	128,000	32,000
`mistral-small-3.1`	Mistral Small 3.1	cloudflare	128,000	8,192
`mistral-nemo`	Mistral Nemo	mistral	128,000	32,000
`ministral-3-14b`	Ministral 3 14B	mistral, bedrock	256,000	32,000
`ministral-3-8b`	Ministral 3 8B	mistral, bedrock	256,000	32,000
`ministral-3-3b`	Ministral 3 3B	mistral, bedrock	256,000	32,000

Alibaba Cloud (16 models)

Qwen 3.7

Model Slug	Name	Providers	Context	Max Output
`qwen3.7-plus`	Qwen3.7 Plus	alibaba	1,000,000	64,000

Qwen 3.6

Model Slug	Name	Providers	Context	Max Output
`qwen3.6-35b`	Qwen3.6 35B A3B	deepinfra	262,144	16,384

Qwen 3.5

Model Slug	Name	Providers	Context	Max Output
`qwen3.5-122b`	Qwen3.5 122B A10B	deepinfra	262,144	32,768
`qwen3.5-397b-a17b`	Qwen3.5 397B A17B	alibaba	256,000	64,000
`qwen3.5-35b-a3b`	Qwen3.5 35B A3B	alibaba	256,000	64,000
`qwen3.5-flash`	Qwen3.5 Flash	alibaba	1,000,000	64,000

Qwen 3 Max (Flagship)

Model Slug	Name	Providers	Context	Max Output
`qwen3-max`	Qwen3 Max	alibaba	256,000	64,000

Qwen 3 Coder (Agentic Coding)

Model Slug	Name	Providers	Context	Max Output
`qwen3-coder-plus`	Qwen3 Coder Plus	alibaba	1,000,000	64,000
`qwen3-coder-flash`	Qwen3 Coder Flash	alibaba	1,000,000	64,000
`qwen3-coder-next`	Qwen3 Coder Next	bedrock	256,000	65,536
`qwen3-coder-30b-a3b`	Qwen3 Coder 30B A3B	bedrock	256,000	65,536

Qwen 3

Model Slug	Name	Providers	Context	Max Output
`qwen3-32b`	Qwen3 32B	bedrock	32,000	8,192
`qwen3-30b`	Qwen3 30B	cloudflare	32,768	8,192
`qwen3-next-80b-a3b`	Qwen3 Next 80B A3B	bedrock	128,000	8,192
`qwen3-vl-235b-a22b`	Qwen3 VL 235B A22B	bedrock	128,000	8,192

QwQ (Reasoning)

Model Slug	Name	Providers	Context	Max Output
`qwq-32b`	QwQ 32B	cloudflare	24,000	16,384

DeepSeek (5 models)

Model Slug	Name	Providers	Context	Max Output
`deepseek-v4-pro`	DeepSeek V4 Pro	deepseek	1,000,000	384,000
`deepseek-v4-flash`	DeepSeek V4 Flash	deepseek	1,000,000	384,000
`deepseek-r1`	DeepSeek R1	bedrock	128,000	32,768
`deepseek-v3-2`	DeepSeek V3.2	vertex, deepinfra, bedrock	163,840	65,536
`deepseek-r1-distill-32b`	DeepSeek R1 Distill 32B	cloudflare	80,000	80,000

z.ai (8 models)

Model Slug	Name	Providers	Context	Max Output
`glm-5`	GLM-5	vertex, zai	200,000	128,000
`glm-5.1`	GLM-5.1	deepinfra, novita	204,800	131,072
`glm-4.7`	GLM-4.7	zai, bedrock	200,000	128,000
`glm-4.7-flash`	GLM-4.7 Flash	bedrock, novita, cloudflare	200,000	131,072
`glm-4.6`	GLM-4.6	zai	200,000	128,000
`glm-4.6v`	GLM-4.6v	zai	131,072	32,768
`glm-4.5`	GLM-4.5	zai	128,000	96,000
`glm-4.5v`	GLM-4.5v	zai	131,072	16,384

Amazon (5 models)

Model Slug	Name	Providers	Context	Max Output
`nova-premier`	Amazon Nova Premier	bedrock	1,000,000	20,000
`nova-pro`	Amazon Nova Pro	bedrock	300,000	5,000
`nova-lite`	Amazon Nova Lite	bedrock	300,000	5,000
`nova-2-lite`	Amazon Nova 2 Lite	bedrock	256,000	5,000
`nova-micro`	Amazon Nova Micro	bedrock	128,000	5,000

MiniMax (7 models)

Model Slug	Name	Providers	Context	Max Output
`minimax-m2-7`	MiniMax M2.7	minimax	204,800	131,072
`minimax-m2-7-highspeed`	MiniMax M2.7 Highspeed	minimax	204,800	131,072
`minimax-m2-5`	MiniMax M2.5	minimax	204,800	8,192
`minimax-m2-5-highspeed`	MiniMax M2.5 Highspeed	minimax	204,800	8,192
`minimax-m2-1`	MiniMax M2.1	minimax, bedrock	204,800	8,192
`minimax-m2-1-highspeed`	MiniMax M2.1 Highspeed	minimax	204,800	8,192
`minimax-m2`	MiniMax M2	minimax, bedrock	204,800	8,192

Moonshot AI (3 models)

Model Slug	Name	Providers	Context	Max Output
`kimi-k2-thinking`	Kimi K2 Thinking	bedrock	256,000	65,535
`kimi-k2-6`	Kimi K2.6	deepinfra, cloudflare	262,144	262,144
`kimi-k2-5`	Kimi K2.5	bedrock	256,000	33,000

NVIDIA (2 models)

Model Slug	Name	Providers	Context	Max Output
`nemotron-3-120b-a12b`	Nemotron 3 120B A12B	cloudflare	256,000	256,000
`nemotron-3-nano-omni-30b-a3b-reasoning`	Nemotron 3 Nano Omni 30B A3B Reasoning	deepinfra	131,072	16,384

Xiaomi (2 models)

Model Slug	Name	Providers	Context	Max Output
`mimo-v2.5-pro`	MiMo V2.5 Pro	deepinfra, novita	1,048,576	131,072
`mimo-v2.5`	MiMo V2.5	deepinfra	1,000,000	16,384

Cohere (2 models)

Model Slug	Name	Providers	Context	Max Output
`command-a`	Command A	cohere	256,000	8,192
`command-a-vision`	Command A Vision	cohere	128,000	8,000

Writer (2 models)

Model Slug	Name	Providers	Context	Max Output
`palmyra-x5`	Palmyra X5	bedrock	128,000	8,192
`palmyra-x4`	Palmyra X4	bedrock	128,000	8,192

AI21 Labs (2 models)

Model Slug	Name	Providers	Context	Max Output
`jamba-1-5-large`	Jamba 1.5 Large	bedrock	256,000	4,096
`jamba-1-5-mini`	Jamba 1.5 Mini	bedrock	256,000	4,096

StepFun AI (1 model)

Model Slug	Name	Providers	Context	Max Output
`step-3-5-flash`	Step 3.5 Flash	deepinfra	262,100	16,400

IBM (1 model)

Model Slug	Name	Providers	Context	Max Output
`ibm-granite-micro`	IBM Granite Micro	cloudflare	131,000	4,096

Providers

Models are available across 17 providers. The same model may be offered by multiple providers with different pricing, latency, and feature support.

Provider	Slug	Description
OpenAI	`openai`	Direct OpenAI API access
Anthropic	`anthropic`	Direct Anthropic API access (ZDR available)
Azure	`azure`	Microsoft Azure OpenAI Service
AWS Bedrock	`bedrock`	Amazon Bedrock managed inference
Google Vertex AI	`vertex`	Google Cloud Vertex AI
Google AI Studio	`ai-studio`	Google AI Studio
xAI	`xai`	xAI direct API access
Cohere	`cohere`	Cohere direct API access
Mistral	`mistral`	Mistral AI direct API access
Alibaba Cloud	`alibaba`	Alibaba Cloud Model Studio (DashScope) direct API access
Cloudflare	`cloudflare`	Cloudflare Workers AI
z.ai	`zai`	Zhipu AI direct API access
MiniMax	`minimax`	MiniMax direct API access
DeepSeek	`deepseek`	DeepSeek direct API access
DeepInfra	`deepinfra`	DeepInfra inference platform
Novita	`novita`	Novita AI inference platform
Blue Lobster	`bluelobster`	Blue Lobster inference

Model Selection

There are three ways to specify which model to use:

Method	Format	Example	Behavior
Model slug	`"model-slug"`	`"claude-opus-4-7"`	Auto-routes to the best provider
Provider-pinned	`"provider/model-slug"`	`"anthropic/claude-opus-4-7"`	Uses specific provider, falls back to others
Auto	`"auto"`	`"auto"`	System selects optimal model and provider

See Routing for full details on provider selection, fallback, and optimization metrics.

Querying Models Programmatically

Use the Models API to get real-time model data including current pricing:

# List all models
curl https://api.concentrate.ai/v1/models \
  -H "Authorization: Bearer YOUR_API_KEY"

# Get a specific model
curl https://api.concentrate.ai/v1/models/claude-opus-4-7 \
  -H "Authorization: Bearer YOUR_API_KEY"

# List models by author
curl https://api.concentrate.ai/v1/models/authors/anthropic \
  -H "Authorization: Bearer YOUR_API_KEY"

# List models by provider
curl https://api.concentrate.ai/v1/models/providers/bedrock/models \
  -H "Authorization: Bearer YOUR_API_KEY"

List Models

Full API reference for querying models

Routing

How provider selection and fallback works

Zero Data Retention

ZDR-certified models and providers

Quickstart

Make your first API call

​Overview

​Models by Author

​OpenAI (23 models)

​Frontier

​Codex (Agentic Coding)

​Reasoning

​Previous Generation

​Open-Weight (GPT-OSS)

​Anthropic (10 models)

​Current Generation

​Previous Generation

​Google (10 models)

​Gemini

​Gemma (Open-Weight)

​xAI (13 models)

​Grok 4.20

​Grok 4 & 4.x

​Grok 3

​Code

​Meta (12 models)

​Llama 4

​Llama 3.x

​Llama 3

​Mistral (13 models)

​Frontier

​Magistral (Reasoning)

​Code

​Small & Efficient

​Alibaba Cloud (16 models)

​Qwen 3.7

​Qwen 3.6

​Qwen 3.5

​Qwen 3 Max (Flagship)

​Qwen 3 Coder (Agentic Coding)

​Qwen 3

​QwQ (Reasoning)

​DeepSeek (5 models)

​z.ai (8 models)

​Amazon (5 models)

​MiniMax (7 models)

​Moonshot AI (3 models)

​NVIDIA (2 models)

​Xiaomi (2 models)

​Cohere (2 models)

​Writer (2 models)

​AI21 Labs (2 models)

​StepFun AI (1 model)

​IBM (1 model)

​Providers

​Model Selection

​Querying Models Programmatically

​Related Documentation

List Models

Routing

Zero Data Retention

Quickstart

Overview

Models by Author

OpenAI (23 models)

Frontier

Codex (Agentic Coding)

Reasoning

Previous Generation

Open-Weight (GPT-OSS)

Anthropic (10 models)

Current Generation

Previous Generation

Google (10 models)

Gemini

Gemma (Open-Weight)

xAI (13 models)

Grok 4.20

Grok 4 & 4.x

Grok 3

Code

Meta (12 models)

Llama 4

Llama 3.x

Llama 3

Mistral (13 models)

Frontier

Magistral (Reasoning)

Code

Small & Efficient

Alibaba Cloud (16 models)

Qwen 3.7

Qwen 3.6

Qwen 3.5

Qwen 3 Max (Flagship)

Qwen 3 Coder (Agentic Coding)

Qwen 3

QwQ (Reasoning)

DeepSeek (5 models)

z.ai (8 models)

Amazon (5 models)

MiniMax (7 models)

Moonshot AI (3 models)

NVIDIA (2 models)

Xiaomi (2 models)

Cohere (2 models)

Writer (2 models)

AI21 Labs (2 models)

StepFun AI (1 model)

IBM (1 model)

Providers

Model Selection

Querying Models Programmatically

Related Documentation