54 models. One key. No per-token bill.

English text embeddings, base (768 dimensions).

Context

512

Output

—

Input

Text

English text embeddings, large (1024 dimensions).

Context

512

Output

—

Input

Text

Multilingual, multi-granularity text embeddings (1024 dimensions).

Context

60K

Output

—

Input

Text

English text embeddings, small and fast (384 dimensions).

Context

512

Output

—

Input

Text

Whisper speech recognition, multilingual.

Context

—

Output

—

Input

Audio

Whisper large-v3-turbo speech recognition, multilingual.

Context

—

Output

—

Input

Audio

Whisper tiny, English-only speech recognition. Fastest.

Context

—

Output

—

Input

Audio

Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family.

• Reasoning
• Image input
• File input
• Tool calling

Context

1M

Output

128K

Input

Text, Images, Files

Claude Opus 5 is Anthropic’s flagship model for demanding reasoning, coding, and long-horizon agentic work.

• Reasoning
• Image input
• File input
• Tool calling

Context

1M

Output

128K

Input

Text, Images, Files

Sonnet 5 is Anthropic's most capable Sonnet-class model, with frontier performance across coding, agents, and professional work.

• Reasoning
• Image input
• File input
• Tool calling

Context

1M

Output

128K

Input

Text, Images, Files

DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total.

• Reasoning
• Tool calling
• JSON output

Context

1M

Output

944K

Input

Text

DeepSeek’s DeepSeek V4 Flash Vision Exp is a reasoning-capable language model with a 1M-token context window, supporting image input, tool calling and json output.

• Reasoning
• Image input
• Tool calling
• JSON output

Context

1M

Output

262K

Input

Text, Images

DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek. This is the GA release of DeepSeek V4 Pro.

• Reasoning
• Tool calling
• JSON output

Context

1M

Output

393K

Input

Text

DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture.

• Reasoning
• Image input
• Tool calling
• JSON output

Context

1M

Output

944K

Input

Text, Images

Kagiro upstream model deepseek-4-1-flash

Context

—

Output

—

Input

Text

Compact multilingual text embeddings from Google (768 dimensions).

Context

2K

Output

—

Input

Text

Gemini 3.6 Flash is a high-efficiency model from Google for coding, agentic workflows, and web and app development.

• Reasoning
• Image input
• File input
• Tool calling

Context

1M

Output

66K

Input

Text, Images, Video, Files, Audio

Gemini 3.7 Flash is a multimodal model from Google for fast agentic workflows, coding, and complex multi-step reasoning.

• Reasoning
• Image input
• File input
• Tool calling

Context

1M

Output

66K

Input

Text, Images, Video, Files, Audio

Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.

• Reasoning
• Image input
• File input
• Tool calling

Context

1M

Output

66K

Input

Text, Images, Video, Files, Audio

GLM-5.1 delivers a major leap in coding capability, with particularly significant gains in handling long-horizon tasks.

• Reasoning
• Tool calling
• JSON output

Context

205K

Output

128K

Input

Text

GLM 5.2 is a large-scale reasoning model from Z.ai.

• Reasoning
• Tool calling
• JSON output

Context

1M

Output

131K

Input

Text

GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks.

• Reasoning
• Image input
• Tool calling
• JSON output

Context

1M

Output

944K

Input

Text

GLM-5.3-Flash is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks.

• Reasoning
• Image input
• Tool calling
• JSON output

Context

1M

Output

944K

Input

Text, Images, Video

GPT-5.3-Codex is OpenAI’s most advanced agentic coding model, combining the frontier software engineering performance of GPT-5.2-Codex with the broader reasoning and professional knowledge capabilities of GPT-5.2.

• Reasoning
• Image input
• File input
• Tool calling

Context

400K

Output

128K

Input

Text, Images, Files

GPT-5.6 Luna is a fast, cost-efficient model in OpenAI's GPT-5.6 series.

• Reasoning
• Image input
• File input
• Tool calling

Context

1.1M

Output

128K

Input

Files, Images, Text

GPT-5.6 Sol is the flagship model in OpenAI's GPT-5.6 series.

• Reasoning
• Image input
• File input
• Tool calling

Context

1.1M

Output

128K

Input

Files, Images, Text

GPT-5.6 Terra is a balanced model in OpenAI's GPT-5.6 series, positioned between the flagship Sol tier and the cost-efficient Luna tier.

• Reasoning
• Image input
• File input
• Tool calling

Context

1.1M

Output

128K

Input

Files, Images, Text

GPT-6 Sol is the cost-efficient high-end model in OpenAI's GPT-6 series, positioned below the flagship GPT-6 Astra and above the fast GPT-6 Luna tier.

• Reasoning
• Image input
• File input
• Tool calling

Context

1.1M

Output

128K

Input

Files, Images, Text

Grok 4.6 is a model from SpaceXAI with frontier performance on coding, knowledge work, and STEM. It is succeeded by [Grok 4.7](/x-ai/grok-4.7).

• Reasoning
• Image input
• File input
• Tool calling

Context

500K

Output

450K

Input

Text, Images, Files

Hy3 is a 295B-parameter Mixture-of-Experts model from Tencent (21B active, 192 experts with top-8 routing) built for reasoning, agentic workflows, and real-world production use.

• Reasoning
• Tool calling
• JSON output

Context

262K

Output

128K

Input

Text

Hy3 is a 295B-parameter Mixture-of-Experts model from Tencent (21B active, 192 experts with top-8 routing) built for reasoning, agentic workflows, and real-world production use.

• Reasoning
• Image input
• Tool calling

Context

262K

Output

236K

Input

Text

Tencent: Hy4 preview is a mixture-of-experts model from Tencent, with 49B active parameters out of 770B total.

• Reasoning
• Image input
• File input
• Tool calling

Context

1M

Output

64K

Input

Text

Inkling is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 41B active parameters out of 975B total.

• Reasoning
• Image input
• Tool calling

Context

524K

Output

472K

Input

Text, Images, Audio

jev answers typed questions about a state — yes/no, choice and score — with one calibrated result per question, instead of generating free text.

• Typed decisions

Context

32K

Output

—

Input

Text

jev-latest answers typed questions about a state — yes/no, choice and score — with one calibrated result per question, instead of generating free text.

• Typed decisions

Context

32K

Output

—

Input

Text

Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration.

• Reasoning
• Image input
• Tool calling
• JSON output

Context

262K

Output

236K

Input

Text, Images

MoonshotAI: Kimi K2.7 Code is a coding-focused model in Moonshot AI's Kimi K2 family, built to complete end-to-end programming tasks reliably over long contexts.

• Reasoning
• JSON output

Context

262K

Output

236K

Input

Text, Images

Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI.

• Reasoning
• Image input
• Tool calling
• JSON output

Context

1M

Output

944K

Input

Text, Images, Video

Ling 3.0 Flash VL by Inclusionai with vision and reasoning capabilities via Vercel AI Gateway

• Reasoning
• Image input
• Tool calling
• JSON output

Context

262K

Output

33K

Input

Text, Images, Video

Mercury 2.5 is the fastest reasoning LLM, and the latest diffusion LLM (dLLM) from Inception.

• Reasoning
• Tool calling
• JSON output

Context

260K

Output

66K

Input

Text

MiMo-V2.5 is a native omnimodal model by Xiaomi.

• Reasoning
• Image input
• Tool calling
• JSON output

Context

1.1M

Output

131K

Input

Text, Audio, Images, Video

MiniMax-M2.7 is a next-generation large language model designed for autonomous, real-world productivity and continuous improvement.

• Reasoning
• Image input
• Tool calling
• JSON output

Context

205K

Output

177K

Input

Text

MiniMax-M3 is a multimodal foundation model from MiniMax.

• Reasoning
• Image input
• Tool calling
• JSON output

Context

1M

Output

512K

Input

Text, Images, Video

Muse Spark 1.3 Contributor is the cost-efficient contributor tier of Meta’s multimodal reasoning model for experimentation, learning, and early-stage agentic, multi-agent, and coding workflows.

• Reasoning
• Image input
• File input
• Tool calling

Context

1M

Output

944K

Input

Text, Images, Video, Files

NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications.

• Reasoning
• Tool calling
• JSON output

Context

262K

Output

236K

Input

Text

NVIDIA Nemotron 3.5 Lightning is an open mixture-of-experts model from NVIDIA, with 3B active parameters out of 30B total.

• Reasoning
• Tool calling
• JSON output

Context

262K

Output

131K

Input

Text

Kagiro upstream model opus4-7

• Tool calling

Context

—

Output

—

Input

Text

Multilingual text embeddings from the Qwen3 Embedding series.

Context

8K

Output

—

Input

Text

Qwen3.8 27B is an open-weight dense vision-language model from Qwen.

• Reasoning
• Image input
• Tool calling
• JSON output

Context

1M

Output

131K

Input

Text, Images, Video

Qwen3.8 Flash is a multimodal reasoning model from Alibaba. It is suited for coding assistance, agentic workflows, visual understanding, document and codebase analysis, desktop interaction, chart analysis, and long-video analysis.

• Reasoning
• Image input
• JSON output

Context

1M

Output

131K

Input

Text, Images, Video

Qwen3.8 Max 0902 is an updated snapshot of Qwen3.8 Max from Alibaba's Qwen team.

• Reasoning
• Image input
• Tool calling
• JSON output

Context

1M

Output

131K

Input

Text, Images, Video

Kagiro upstream model qwen3.8-plus

• Tool calling

Context

—

Output

—

Input

Text

Whisper large-v3 on Groq: best accuracy, multilingual.

Context

—

Output

—

Input

Audio

Whisper large-v3-turbo on Groq: very fast, multilingual.

Context

—

Output

—

Input

Audio