One API for every LLM
GPT, Claude, Gemini, Grok and 150+ more behind one endpoint.
RouteLLM API — by Abacus.AI
One API.
Every Frontier LLM.
Access 150+ models — GPT, Claude, Gemini, Grok, DeepSeek, Qwen and more — through a single OpenAI-compatible API. Smart routing picks the best model for every prompt.
$10 $7 for the first month, then $10 billed monthly
— included with ChatLLM Teams
Key Features
Why RouteLLM
Smart routing to the best model
Every prompt is matched to the model that answers it best.
100+ AI Models
Text, image, audio and video models — new releases added as they launch.
Failover & prompt caching
Automatic failover between providers and built-in caching.
The Catalog
Every frontier model. One key.
Ranked by LiveBench score — routed automatically, or pick your own. Click any model for details.
| Model | LiveBench | Context | Modalities | Price /1M | Capabilities |
|---|---|---|---|---|---|
|
RouteLLM · Abacus.AI
RouteLLM is a model that routes the user's message to the appropriate text-generation model. We will route to one of Sonnet 4.6, GPT-5.4, or Gemini 3.1 Flash models. The price shown is for the most expensive model (Sonnet 4.6) that we use.
|
—/auto | — |
|
$2.00 / $10.00 | Smart Router |
|
A
Claude Fable 5 · Anthropic
Claude Fable 5 from Anthropic is the next-generation Claude model, delivering major advances in coding, reasoning, and agentic tasks. It excels in tasks requiring long, complex reasoning chains, but is also expensive.
|
83.0
|
1M |
|
$10.00 / $50.00 | |
|
O
GPT-5.6 Sol · OpenAI
GPT-5.6 Sol from OpenAI is a flagship model with advanced reasoning capabilities and a 1M context window.
|
81.0
|
1M |
|
$5.00 / $30.00 | |
|
O
GPT-5.5 · OpenAI
GPT-5.5 from OpenAI is a flagship model with advanced reasoning capabilities and a 1M context window.
|
80.2
|
1M |
|
$5.00 / $30.00 | |
|
A
Claude Opus 5 · Anthropic
Claude Opus 5 from Anthropic is a highly capable AI model, focused on advanced coding, reasoning, and agentic tasks, with improved performance in multi-file code refactoring, debugging, and detail-oriented reasoning. It excels in tasks requiring long, complex reasoning chains, but is also expensive.
|
80.1
|
1M |
|
$5.00 / $25.00 | |
|
M
Kimi K3 · Moonshot
Kimi K3 is Moonshot AI's flagship open-weight multimodal reasoning model with a 1M-token context window, built for long-horizon coding, agentic workflows, and end-to-end knowledge work.
|
79.2
|
1.0M |
|
$3.00 / $15.00 | |
|
O
GPT-5.4 · OpenAI
GPT-5.4
|
78.0
|
400K |
|
$2.50 / $15.00 | |
|
M
Muse Spark 1.2 · Other
Muse Spark 1.2 is Meta's natively multimodal reasoning model with a 1M token context window and text, image, video, audio, and PDF inputs, co-trained with the Muse Code agent for long-horizon coding, debugging, and whole-repository generation, with native tool use and parallel tool calling.
|
78.0
|
1.0M |
|
$1.25 / $4.25 | |
|
O
GPT-5.6 Terra · OpenAI
GPT-5.6 Terra from OpenAI is a flagship model with advanced reasoning capabilities and a 1M context window.
|
77.9
|
1M |
|
$2.00 / $12.00 | |
|
G
Gemini 3.1 Pro · Google
Gemini 3.1 Pro
|
77.0
|
1.0M |
|
$2.00 / $12.00 | |
|
A
Claude Opus 4.7 · Anthropic
Claude Opus 4.7 from Anthropic is a highly capable AI model, focused on advanced coding, reasoning, and agentic tasks, with improved performance in multi-file code refactoring, debugging, and detail-oriented reasoning. It excels in tasks requiring long, complex reasoning chains, but is also expensive.
|
76.5
|
1M |
|
$5.00 / $25.00 | |
|
A
Claude Opus 4.8 · Anthropic
Claude Opus 4.8 from Anthropic is a highly capable AI model, focused on advanced coding, reasoning, and agentic tasks, with improved performance in multi-file code refactoring, debugging, and detail-oriented reasoning. It excels in tasks requiring long, complex reasoning chains, but is also expensive.
|
76.2
|
1M |
|
$5.00 / $25.00 | |
|
A
Claude Sonnet 5 · Anthropic
Claude Sonnet 5 from Anthropic is a balanced AI model, offering excellent performance in coding, reasoning, and general tasks with improved efficiency. It provides a strong balance between capability and cost, making it ideal for most use cases.
|
76.0
|
1M |
|
$2.00 / $10.00 | |
|
X
Grok 4.5 · xAI
Grok 4.5, the latest frontier multimodal model from xAI with a 500K context window optimized for high-performance agentic tool calling. Reasoning is disabled.
|
75.8
|
500K |
|
$2.00 / $6.00 | |
|
M
Muse Spark 1.1 · Other
Muse Spark 1.1 is Meta's natively multimodal reasoning model with a 1,000,000 token context window, built for agentic tasks with native tool use, parallel tool calling, and strong coding performance.
|
75.3
|
1M |
|
$1.25 / $4.25 | |
|
G
Gemini 3.5 Flash · Google
Gemini 3.5 Flash from Google is an advanced multimodal model designed for high-quality reasoning and efficiency. It offers strong performance across text, image, audio, and video tasks.
|
74.6
|
1.0M |
|
$1.50 / $9.00 | |
|
O
GPT-5.2 · OpenAI
GPT-5.2
|
74.6
|
400K |
|
$1.75 / $14.00 | |
|
A
Claude Opus 4.6 · Anthropic
Claude Opus 4.6 from Anthropic is a highly capable AI model, focused on advanced coding, reasoning, and agentic tasks, with improved performance in multi-file code refactoring, debugging, and detail-oriented reasoning. It excels in tasks requiring long, complex reasoning chains, but is also expensive.
|
74.5
|
1M |
|
$5.00 / $25.00 | |
|
D
Deepseek V4 Flash 0731 · DeepSeek
DeepSeek V4 Flash is an advanced, high-performance large language model (LLM) developed by DeepSeek.
|
74.2
|
1M |
|
$0.14 / $0.28 | |
|
G
Gemini 3.6 Flash · Google
Gemini 3.6 Flash from Google is an advanced multimodal reasoning model that delivers better coding, knowledge work, and multimodal performance with greater token efficiency than Gemini 3.5 Flash.
|
73.6
|
1.0M |
|
$1.50 / $7.50 | |
|
O
GPT-5.6 Luna · OpenAI
GPT-5.6 Luna from OpenAI is a flagship model with advanced reasoning capabilities and a 1M context window.
|
73.6
|
1M |
|
$0.20 / $1.20 | |
|
Z
GLM 5.2 · Z.AI
GLM-5.2 from Z.AI is a flagship model excelling at long-horizon autonomous tasks with a 1M token context window, strong coding and agentic capabilities.
|
73.2
|
1.0M |
|
$1.40 / $4.40 | |
|
A
Claude Sonnet 4.6 · Anthropic
Claude Sonnet 4.6 from Anthropic is a balanced AI model, offering excellent performance in coding, reasoning, and general tasks with improved efficiency. It provides a strong balance between capability and cost, making it ideal for most use cases.
|
73.0
|
1M |
|
$3.00 / $15.00 | |
|
A
Claude Opus 4.5 · Anthropic
Claude Opus 4.5 from Anthropic is an AI model focused on advanced coding, reasoning, and agentic tasks, with improved performance in multi-file code refactoring, debugging, and detail-oriented reasoning. It excels in tasks requiring long, complex reasoning chains, but is also expensive.
|
72.6
|
200K |
|
$5.00 / $25.00 | |
|
T
Tinker Inkling · Other
Inkling is Thinking Machines Lab's first open-weights model, a 975B-parameter mixture-of-experts (41B active) that reasons across text and images, served with a 262,144 token context window. Supports tools and streaming.
|
71.9
|
262K |
|
$3.74 / $9.36 | |
|
M
Kimi K2.6 · Moonshot
Kimi K2.6 is a natively multimodal model with powerful coding capabilities and enhanced agent performance. It builds on the K2.5 foundation with improved reasoning depth, cleaner agent planning, and faster multi-step tool call reliability.
|
70.5
|
262K |
|
$0.95 / $4.00 | |
|
O
GPT-5.4 Nano · OpenAI
GPT-5.4 Nano
|
69.6
|
400K |
|
$0.20 / $1.25 | Text |
|
M
Kimi K2.7 Code · Moonshot
Kimi K2.7 Code is Kimi's most capable coding model to date. It follows instructions more reliably in long contexts and completes coding tasks with higher success rates.
|
68.4
|
262K |
|
$0.95 / $4.00 | |
|
M
MiniMax M3 · MiniMax
MiniMax M3 is MiniMax's latest flagship reasoning model with a 1,000,000 token context window and 131,072 max output tokens. Supports tools, structured output, and reasoning.
|
67.3
|
1M |
|
$0.30 / $1.20 | Text |
|
O
GPT-5.4 Mini · OpenAI
GPT-5.4 Mini
|
66.4
|
400K |
|
$0.75 / $4.50 | |
|
Q
Qwen3.6 27B · Qwen
Qwen3.6 27B is an open-weight dense 27B-parameter model from Alibaba's Qwen3.6 series, delivering flagship-level coding and reasoning performance in a compact package. It supports a 256K context window and is well suited for software development, multilingual tasks, and tool-calling workflows.
|
64.0
|
262K |
|
$0.32 / $3.20 | Text |
|
G
Gemini 3.5 Flash Lite · Google
Gemini 3.5 Flash Lite from Google is the fastest and most cost-effective Gemini 3.5 series model, built for high-throughput, low-latency workloads like agentic search and document processing.
|
63.9
|
1.0M |
|
$0.30 / $2.50 | |
|
X
Grok 4.3 · xAI
Grok 4.3, the latest frontier multimodal model from xAI with a 1M context window optimized for high-performance agentic tool calling. Reasoning is disabled.
|
62.3
|
1M |
|
$1.25 / $2.50 | |
|
A
Claude Haiku 4.5 · Anthropic
Claude Haiku 4.5
|
— | 200K |
|
$1.00 / $5.00 | |
|
D
Deepseek V4 Pro · DeepSeek
DeepSeek V4 Pro is an advanced, high-performance large language model (LLM) developed by DeepSeek.
|
— | 1M |
|
$1.32 / $3.96 | |
|
G
Gemini 3 Flash · Google
Gemini 3 Flash from Google is an advanced large language model. It is designed for efficiency while balancing reasoning capabilities.
|
— | 1.0M |
|
$0.50 / $3.00 | |
|
G
Gemini 3.1 Flash Lite · Google
Gemini 3.1 Flash Lite from Google is a fast and cost-efficient Gemini 3 series model, built for high-volume developer workloads at scale.
|
— | 1.0M |
|
$0.25 / $1.50 | |
|
G
Gemini 3.7 Flash · Google
Gemini 3.7 Flash from Google is an advanced multimodal reasoning model built for coding and agentic workflows that delivers better debugging, knowledge work, and multimodal performance with tunable thinking levels than Gemini 3.6 Flash.
|
— | 1.0M |
|
$0.75 / $3.75 | |
|
X
Grok 4.6 · xAI
Grok 4.6, the latest frontier multimodal model from xAI with a 500K context window built for coding, agentic tasks, and knowledge work. Reasoning is disabled.
|
— | 500K |
|
$2.00 / $6.00 | |
|
A
Claude Sonnet 4.5 · Anthropic
Claude Sonnet 4.5 advances over its predecessor, Sonnet 4, delivering stronger coding and reasoning performance with greater precision and control. Makes agentic programming better. It excels in tasks requiring long, complex reasoning chains, but is also expensive.
|
— | 200K |
|
$3.00 / $15.00 | |
|
D
Deepseek V3.2 · DeepSeek
DeepSeek-V3.2-Exp is an intermediate step toward the next-generation architecture of the DeepSeek models by introducing DeepSeek Sparse Attention - a sparse attention mechanism designed to explore and validate optimizations for training and inference efficiency in long-context scenarios.
|
— | 131K |
|
$0.27 / $0.40 | Text |
|
G
Gemini 2.5 Flash · Google
Gemini 2.5 Flash is a large multimodal language model from Google. It is designed for speed, efficiency, and cost-effectiveness in AI tasks that require high volume and frequency.
|
— | 1.0M |
|
$0.30 / $2.50 | |
|
G
Gemini 2.5 Pro · Google
Gemini 2.5 Pro is Google's advanced large language model. It is designed for complex tasks that require deep reasoning, advanced coding, and multimodal understanding.
|
— | 1.0M |
|
$1.25 / $10.00 | |
|
G
Gemma 4 31B IT · Google
Gemma 4 31B IT is Google's open-weight vision-language model with a 256K context window. It supports interleaved text and image inputs, structured output, function calling, and reasoning tasks.
|
— | 262K |
|
$0.14 / $0.40 | Text |
|
Z
GLM 4.5 · Z.AI
GLM-4.5 from Z.AI is a flagship model, purpose-built for agent-based applications. It leverages a Mixture-of-Experts (MoE) architecture and supports a context length of up to 128k tokens. GLM-4.5 delivers significantly enhanced capabilities in reasoning, code generation, and agent alignment. It supports a hybrid inference mode with two options, a "thinking mode" designed for complex reasoning and tool use, and a "non-thinking mode" optimized for instant responses.
|
— | 131K |
|
$0.60 / $2.20 | Text |
|
Z
GLM 4.6 · Z.AI
GLM-4.6 from Z.AI is a flagship model with a 128k context window, better coding and agentic performance and more refined writing
|
— | 203K |
|
$0.60 / $2.20 | Text |
|
Z
GLM 4.7 · Z.AI
GLM-4.7 from Z.AI is a flagship model featuring enhanced programming capabilities and stable multi-step reasoning/execution. It demonstrates significant improvements in executing complex agent tasks while delivering more natural conversational experiences and superior front-end aesthetics.
|
— | 205K |
|
$0.60 / $2.20 | Text |
|
Z
GLM 5 · Z.AI
GLM-5 from Z.AI is a flagship model featuring enhanced capabilities in reasoning, code generation, and agent alignment.
|
— | 205K |
|
$1.00 / $3.20 | Text |
|
Z
GLM 5.1 · Z.AI
GLM-5.1 from Z.AI is a flagship model featuring enhanced capabilities in reasoning, code generation, and agent alignment.
|
— | 205K |
|
$1.40 / $4.40 | Text |
|
O
GPT 4o Mini Transcribe · OpenAI
GPT 4o Mini Transcribe
|
— | 16K |
|
$1.25 / $5.00 | Text |
|
O
GPT 4o Transcribe · OpenAI
GPT 4o Transcribe
|
— | 16K |
|
$2.50 / $10.00 | Text |
|
O
GPT Realtime Whisper · OpenAI
GPT Realtime Whisper
|
— | 16K |
|
$283.33 / $0.00 | Text |
|
O
GPT Transcribe · OpenAI
GPT Transcribe
|
— | 16K |
|
$75.00 / $0.00 | Text |
|
O
GPT-4.1 · OpenAI
GPT-4.1 is designed to be faster, cheaper, and more reliable for specific tasks compared to the o-series models, especially for coding and tool-based operations.
|
— | 1.0M |
|
$2.00 / $8.00 | Text |
|
O
GPT-4.1 Mini · OpenAI
GPT-4.1 Mini is a smaller, faster, and cheaper version of GPT-4.1, designed for lightweight tasks and lower latency requirements.
|
— | 1.0M |
|
$0.40 / $1.60 | Text |
|
O
GPT-4.1 Nano · OpenAI
GPT-4.1 Nano is the smallest and cheapest model in the GPT-4.1 family, ideal for basic tasks and low-resource applications.
|
— | 1.0M |
|
$0.10 / $0.40 | Text |
|
O
GPT-4o · OpenAI
GPT-4o from OpenAI is an advanced multimodal AI model that can seamlessly process and generate content across text, images, and audio. This allows for more natural human-computer interaction, as users can engage with it through talking, typing, showing images, and even videos.
|
— | 128K |
|
$2.50 / $10.00 | Text |
|
O
GPT-4o Mini · OpenAI
GPT-4o mini is a highly efficient and cost-effective small language model. t excels in instruction following, multimodal reasoning, and supports a 128k context window.
|
— | 128K |
|
$0.15 / $0.60 | Text |
|
O
GPT-5 · OpenAI
GPT-5 from OpenAI is an advanced model, delivering significant gains in reasoning, code generation quality, and overall user experience. It is a very good model for coding and agentic tasks across industries.
|
— | 400K |
|
$1.25 / $10.00 | Text |
|
O
GPT-5 Mini · OpenAI
GPT-5 Mini is a streamlined variant of GPT-5 built for lighter reasoning workloads. It delivers GPT-5 instruction adherence and safety tuning while offering lower latency and cost.
|
— | 400K |
|
$0.25 / $2.00 | Text |
|
O
GPT-5 Nano · OpenAI
GPT-5 Nano is the fastest and cheapest model in the GPT-5 family. It is a good model for summarization and classification tasks.
|
— | 400K |
|
$0.05 / $0.40 | Text |
|
O
GPT-5.1 · OpenAI
GPT-5.1
|
— | 400K |
|
$1.25 / $10.00 | |
|
O
GPT-5.3 Codex · OpenAI
GPT-5.3 Codex
|
— | 400K |
|
$1.75 / $14.00 | Text |
|
O
GPT-OSS 120B · OpenAI
GPT-OSS from OpenAI is a powerful, open-weight language model designed for high-reasoning, agentic tasks. It supports advanced capabilities like configurable reasoning effort levels, full chain-of-thought access, and native tool use, including function calling, browsing, and structured output generation.
|
— | 128K |
|
$0.08 / $0.44 | Text |
|
M
Kimi K2 Turbo · Moonshot
Context length 256k. High-speed version of kimi-k2, always aligned with the latest kimi-k2 (kimi-k2-0905-preview). Same model parameters as kimi-k2, output speed up to 60 tokens/sec (max 100 tokens/sec)
|
— | 262K |
|
$0.15 / $8.00 | |
|
M
Llama 3.1 8B · Meta
Llama 3.1 8B is a powerful, multilingual open-source large language model developed by Meta. It is very fast and cheap.
|
— | 131K |
|
$0.02 / $0.05 | Text |
|
M
Llama 3.3 70B · Meta
Llama 3.3 70B
|
— | 131K |
|
$0.59 / $0.79 | Text |
|
M
Llama 4 Maverick · Meta
Llama 4 Maverick is a is an advanced, natively multimodal large language model (LLM) developed by Meta. It delivers strong performance across tasks like natural language processing, coding, image recognition, and multimodal reasoning.
|
— | 1.0M |
|
$0.14 / $0.59 | Text |
|
X
MiMo V2 Pro · Xiaomi
Xiaomi MiMo V2 Pro is Xiaomi's flagship reasoning model with a 1M token context window and 128K max output. Supports reasoning via <think> tags, tool use, structured output, and speculative decoding for fast inference.
|
— | 1.0M |
|
$1.00 / $3.00 | Text |
|
M
MiniMax M2.7 · MiniMax
MiniMax M2.7 is MiniMax's flagship text model with a 204,800 token context window. ~60 tokens/sec output. Supports tools, streaming, and JSON mode.
|
— | 205K |
|
$0.30 / $1.20 | Text |
|
G
Nano Banana (Gemini 2.5 Flash Image) · Google
Nano Banana (Gemini 2.5 Flash Image)
|
— | 33K |
|
$0.30 / $30.00 | Text |
|
G
Nano Banana (Gemini 3 Pro Image) · Google
Nano Banana (Gemini 3 Pro Image)
|
— | 66K |
|
$2.00 / $12.00 | Text |
|
G
Nano Banana 2 (Gemini 3.1 Flash Image) · Google
Nano Banana 2 (Gemini 3.1 Flash Image)
|
— | 1.0M |
|
$0.50 / $3.00 | Text |
|
O
o3 · OpenAI
O3 is advance reasoning models, designed for tackling complex tasks requiring deep analytical thinking and problem-solving.
|
— | 200K |
|
$2.00 / $8.00 | Text |
|
O
o3 Mini · OpenAI
o3 Mini
|
— | 200K |
|
$1.10 / $4.40 | Text |
|
O
o3 Pro · OpenAI
o3 Pro
|
— | 200K |
|
$20.00 / $40.00 | Text |
|
O
o4 Mini · OpenAI
OpenAI o4-mini is a compact, efficient, and cost-effective model in the OpenAI o-series, which emphasizes reasoning capabilities.
|
— | 200K |
|
$1.10 / $4.40 | Text |
|
Q
Qwen3 32B · Qwen
Qwen3-32B is a dense 32.8 billion parameter model by Alibaba. The model shows strong performance in reasoning tasks and is optimized for agentic applications, supporting tool-calling and external tool integration.
|
— | 131K |
|
$0.09 / $0.29 | Text |
|
Q
Qwen3 Coder · Qwen
Qwen3-Coder is a powerful, open-source, agentic coding model by Alibaba, known for its great ability to generate code from natural language, debug code, and interact with tools
|
— | 262K |
|
$0.29 / $1.20 | Text |
|
Q
Qwen3.7 Max · Qwen
Qwen3.7-Max is Alibaba's flagship model for the agent era, featuring upgraded reasoning and coding capabilities aimed at long-running agent workloads. It supports a 1M token context window and excels at programming, productivity tasks, and autonomous multi-step execution with tool calling.
|
— | 1M |
|
$2.50 / $7.50 | |
|
Q
Qwen3.8 Max · Qwen
Qwen3.8-Max is Alibaba's flagship 2.4-trillion-parameter MoE model with native multimodal understanding and a 1M token context window. It excels at long-horizon agentic tasks, advanced coding, and complex reasoning with configurable thinking effort.
|
— | 1M |
|
$2.00 / $6.00 | |
|
O
DALL-E · OpenAI
DALL-E
|
— | — |
|
— | Image |
|
B
Dreamina · ByteDance
Dreamina
|
— | — |
|
from $0.03 | Image |
|
B
FLUX 1.1 [pro] · Black Forest Labs
FLUX 1.1 [pro]
|
— | — |
|
from $0.04 | Image |
|
B
FLUX 1.1 [pro] Canny [Edit] · Black Forest Labs
FLUX 1.1 [pro] Canny [Edit]
|
— | — |
|
from $0.05 | Image |
|
B
FLUX 1.1 [pro] Depth [Edit] · Black Forest Labs
FLUX 1.1 [pro] Depth [Edit]
|
— | — |
|
from $0.05 | Image |
|
B
FLUX 1.1 [pro] Ultra · Black Forest Labs
FLUX 1.1 [pro] Ultra
|
— | — |
|
from $0.06 | Image |
|
B
FLUX.1 Kontext · Black Forest Labs
FLUX.1 Kontext
|
— | — |
|
from $0.04 | Image |
|
B
FLUX.1 Kontext [Edit] · Black Forest Labs
FLUX.1 Kontext [Edit]
|
— | — |
|
from $0.04 | Image |
|
B
FLUX.2 · Black Forest Labs
FLUX.2
|
— | — |
|
— | Image |
|
B
FLUX.2 [Pro] · Black Forest Labs
FLUX.2 [Pro]
|
— | — |
|
from $0.03 | Image |
|
O
GPT Image 1.5 · OpenAI
GPT Image 1.5
|
— | — |
|
— | Image |
|
O
GPT Image 2 · OpenAI
GPT Image 2
|
— | — |
|
— | Image |
|
O
GPT Image 2 [Edit] · OpenAI
GPT Image 2 [Edit]
|
— | — |
|
— | Image |
|
O
GPT Image [Edit] · OpenAI
GPT Image [Edit]
|
— | — |
|
— | Image |
|
X
Grok Imagine Image · xAI
Grok Imagine Image
|
— | — |
|
from $0.02 | Image |
|
X
Grok Imagine Image 2 · xAI
Grok Imagine Image 2
|
— | — |
|
from $0.04 | Image |
|
X
Grok Imagine Quality · xAI
Grok Imagine Quality
|
— | — |
|
from $0.05 | Image |
|
T
Hunyuan Image 3.0 · Tencent
Hunyuan Image 3.0
|
— | — |
|
from $0.10 | Image |
|
I
Ideogram 3.0 · Ideogram
Ideogram 3.0
|
— | — |
|
from $0.06 | Image |
|
I
Ideogram Character · Ideogram
Ideogram Character
|
— | — |
|
from $0.10 | Image |
|
I
Imagineart 1.5 · ImagineArt
Imagineart 1.5
|
— | — |
|
from $0.03 | Image |
|
M
Magnific Upscaler · Magnific
Magnific Upscaler
|
— | — |
|
— | Image |
|
M
Midjourney · Midjourney
Midjourney
|
— | — |
|
from $0.04 | Image |
|
G
Nano Banana · Google
Nano Banana
|
— | — |
|
— | Image |
|
G
Nano Banana 2 · Google
Nano Banana 2
|
— | — |
|
from $0.06 | Image |
|
G
Nano Banana Lite · Google
Nano Banana Lite
|
— | — |
|
from $0.03 | Image |
|
G
Nano Banana Pro · Google
Nano Banana Pro
|
— | — |
|
from $0.15 | Image |
|
Q
Qwen Image Edit · Qwen
Qwen Image Edit
|
— | — |
|
from $0.03 | Image |
|
R
Recraft · Recraft
Recraft
|
— | — |
|
from $0.04 | Image |
|
R
Recraft SVG · Recraft
Recraft SVG
|
— | — |
|
from $0.08 | Image |
|
R
Recraft Vectorize · Recraft
Recraft Vectorize
|
— | — |
|
from $0.01 | Image |
|
B
Seedream 4.5 · ByteDance
Seedream 4.5
|
— | — |
|
from $0.04 | Image |
|
B
Seedream 5 Lite · ByteDance
Seedream 5 Lite
|
— | — |
|
— | Image |
|
B
Seedream 5 Pro · ByteDance
Seedream 5 Pro
|
— | — |
|
— | Image |
|
A
Wan 2.7 · Alibaba
Wan 2.7
|
— | — |
|
from $0.03 | Image |
|
B
FLUX 3 · Black Forest Labs
FLUX 3
|
— | — |
|
from $0.06 | Video |
|
G
Gemini Omni Flash · Google
Gemini Omni Flash
|
— | — |
|
— | Video |
|
X
Grok Imagine Video · xAI
Grok Imagine Video
|
— | — |
|
from $0.05 | Video |
|
X
Grok Imagine Video 1.5 · xAI
Grok Imagine Video 1.5
|
— | — |
|
from $0.08 | Video |
|
M
Hailuo 2 · MiniMax
Hailuo 2
|
— | — |
|
from $0.27 | Video |
|
T
Hunyuan Video · Tencent
Hunyuan Video
|
— | — |
|
from $0.40 | Video |
|
K
Kling AI O1 · Other
Kling AI O1
|
— | — |
|
from $0.42 | Video |
|
K
Kling AI O3 · Other
Kling AI O3
|
— | — |
|
from $0.28 | Video |
|
K
Kling AI v1.6 · Other
Kling AI v1.6
|
— | — |
|
— | Video |
|
K
Kling AI v2 · Other
Kling AI v2
|
— | — |
|
from $1.40 | Video |
|
K
Kling AI v2.1 · Other
Kling AI v2.1
|
— | — |
|
from $1.40 | Video |
|
K
Kling AI v2.5 · Other
Kling AI v2.5
|
— | — |
|
from $0.21 | Video |
|
K
Kling AI v2.6 · Other
Kling AI v2.6
|
— | — |
|
from $0.07 | Video |
|
K
Kling AI v3 · Other
Kling AI v3
|
— | — |
|
— | Video |
|
K
Kling v2.6 Motion Control · Other
Kling v2.6 Motion Control
|
— | — |
|
— | Video |
|
K
Kling v3 Motion Control · Other
Kling v3 Motion Control
|
— | — |
|
— | Video |
|
L
Luma Labs · Other
Luma Labs
|
— | — |
|
from $0.40 | Video |
|
M
MiniMax H3 · MiniMax
MiniMax H3
|
— | — |
|
from $0.08 | Video |
|
R
Runway · Other
Runway
|
— | — |
|
from $0.25 | Video |
|
B
Seedance · ByteDance
Seedance
|
— | — |
|
from $0.18 | Video |
|
B
Seedance 1.5 Pro · ByteDance
Seedance 1.5 Pro
|
— | — |
|
from $0.26 | Video |
|
B
Seedance 2.0 · ByteDance
Seedance 2.0
|
— | — |
|
— | Video |
|
B
Seedance 2.0 Mini · ByteDance
Seedance 2.0 Mini
|
— | — |
|
— | Video |
|
B
Seedance 2.5 · ByteDance
Seedance 2.5
|
— | — |
|
— | Video |
|
B
Seedance Pro · ByteDance
Seedance Pro
|
— | — |
|
from $0.74 | Video |
|
S
Sora 2 · Other
Sora 2
|
— | — |
|
from $0.10 | Video |
|
T
Topaz Upscaler · Other
Topaz Upscaler
|
— | — |
|
from $0.10 | Video |
|
V
Veo 3.1 · Other
Veo 3.1
|
— | — |
|
from $1.20 | Video |
|
V
Veo 3.1 Lite · Other
Veo 3.1 Lite
|
— | — |
|
from $0.07 | Video |
|
A
Wan 2.2 · Alibaba
Wan 2.2
|
— | — |
|
from $0.08 | Video |
|
A
Wan 2.5 · Alibaba
Wan 2.5
|
— | — |
|
from $0.05 | Video |
|
A
Wan 2.7 · Alibaba
Wan 2.7
|
— | — |
|
from $0.10 | Video |
|
E
ElevenLabs · ElevenLabs
ElevenLabs
|
— | — |
|
— | Audio |
|
G
Gemini 2.5 Flash TTS · Google
Gemini 2.5 Flash TTS is a dedicated text-to-speech model from Google, supporting audio output generation via the Gemini API.
|
— | 1.0M |
|
$0.50 / $0.00 | Audio |
|
G
Gemini 2.5 Pro TTS · Google
Gemini 2.5 Pro TTS is a high-quality dedicated text-to-speech model from Google, supporting audio output generation via the Gemini API.
|
— | 2.1M |
|
$1.00 / $0.00 | Audio |
|
O
GPT Audio 1.5 · OpenAI
GPT Audio 1.5 supports audio input and output via the chat completions API, enabling speech understanding and generation in a single model call.
|
— | 128K |
|
$2.50 / $10.00 | Audio |
|
O
GPT Audio Mini · OpenAI
GPT Audio Mini is a cost-efficient model supporting audio input and output via the chat completions API, ideal for high-volume speech applications.
|
— | 128K |
|
$0.60 / $2.40 | Audio |
|
H
Hume · Hume
Hume
|
— | — |
|
— | Audio |
|
M
MiniMax Speech 2.8 HD · MiniMax
MiniMax Speech 2.8 HD
|
— | — |
|
— | Audio |
|
O
OpenAI · OpenAI
OpenAI
|
— | — |
|
— | Audio |
|
S
Seed Audio 1.0 · Other
Seed Audio 1.0
|
— | — |
|
— | Audio |
|
S
Seed Speech · Other
Seed Speech
|
— | — |
|
— | Audio |
|
V
VibeVoice · Other
VibeVoice
|
— | — |
|
— | Audio |
No models match your search.
LiveBench scores from livebench.ai (June 2026)
Get Started
How to use it
1
Sign up / Sign in
Create your Abacus.AI account in under a minute.
2
Subscribe to ChatLLM Teams
$7 for your first month (then $10/mo) unlocks the entire model catalog.
3
Grab your API key
Point your OpenAI SDK at our base URL and ship.
quickstart.py
from openai import OpenAI
client = OpenAI(
base_url="https://routellm.abacus.ai/v1",
api_key="YOUR_API_KEY")
r = client.chat.completions.create(
model="route-llm",
messages=[{"role": "user", "content": "Hello!"}])Works with Claude Code, Codex CLI and Cursor
Sign up to ChatLLM to proceed
Get more access to ChatLLM and unlock powerful AI Agent capabilities
Access to 100+ AI models including Fable 5, GPT 5.6 Sol and Seedream 2.0
Get Started
1st Month Discount
First month then, $10/month
100+ AI & Image Models
Vibe Code Apps
General Purpose Agent
CLI + CoWork
SuperComputer
Already a user? Log in here
An AI Super Assistant For Enterprises And Professionals
Copyright © 2026 Abacus.AI. All Rights Reserved