Models

Models available through the gateway, with current pricing.

Provider
Claude
claude-fable-5-30%Reasoning

Anthropic’s newest flagship generation with a 1M-token context — deep multi-step reasoning, agentic autonomy and long-session coherence beyond Opus 4.6.

by AnthropicJun 20261M context$7 in$10$35 out$50
Claude
claude-haiku-4-5-30%Text

Fast, inexpensive Claude model for high-volume tasks: classification, extraction, summarisation and latency-sensitive chat.

by AnthropicOct 2025200K context$0.7 in$1$3.5 out$5
Claude
claude-opus-4-5-30%Reasoning

Opus-tier Claude model for the hardest reasoning, agentic and coding workloads.

by AnthropicNov 2025200K context$3.5 in$5$17.5 out$25
Claude
claude-opus-4-6-30%Reasoning

Anthropic’s most capable 4.x model for deep reasoning, agentic workflows and long-horizon coding tasks, with 1M context.

by AnthropicFeb 20261M context$3.5 in$5$17.5 out$25
Claude
claude-opus-4-7-30%Reasoning

Opus-tier Claude model for the hardest reasoning, agentic and coding workloads.

by AnthropicApr 20261M context$3.5 in$5$17.5 out$25
Claude
claude-opus-4-8-30%Reasoning

Opus-tier Claude model for the hardest reasoning, agentic and coding workloads.

by AnthropicMay 20261M context$3.5 in$5$17.5 out$25
Claude
claude-opus-5-30%Reasoning

Next-generation Opus with a major jump in planning, tool orchestration and hard-problem reasoning. The premium tier of the Claude family.

by AnthropicMay 20261M context$3.5 in$5$17.5 out$25
Claude
claude-sonnet-4-5-30%Text

Sonnet-tier Claude model balancing capability and cost for production workloads.

by AnthropicSep 2025200K context$2.1 in$3$10.5 out$15
Claude
claude-sonnet-4-6-30%Text

Balanced Claude model for production workloads — near-Opus quality on everyday tasks at a fraction of the cost, with strong tool use.

by AnthropicFeb 20261M context$2.1 in$3$10.5 out$15
Claude
claude-sonnet-5-30%Text

Fifth-generation Sonnet: near-Opus capability at $2/$10 — the new price-performance sweet spot of the Claude family.

by AnthropicMar 20261M context$1.4 in$2$7 out$10
DeepSeek
deepseek-chat-30%Text

DeepSeek general chat model — remarkable quality-per-dollar for chat, RAG and structured extraction.

by DeepSeekDec 2025128K context$0.196 in$0.28$0.294 out$0.42
DeepSeek
deepseek-coder-30%Coding

DeepSeek coding model for completion, generation and repair.

by DeepSeek128K context$0.098 in$0.14$0.196 out$0.28
DeepSeek
deepseek-r1-30%Reasoning

DeepSeek reasoning model exposing chain-of-thought at very low cost.

by DeepSeekJan 202564K context$0.385 in$0.55$1.533 out$2.19
DeepSeek
deepseek-reasoner-30%Reasoning

DeepSeek reasoning model exposing chain-of-thought for math, code and analysis at very low cost.

by DeepSeekDec 2025128K context$0.196 in$0.28$0.294 out$0.42
DeepSeek
deepseek-v3.2-30%Text

DeepSeek model with remarkable quality-per-dollar for chat and RAG.

by DeepSeekSep 2025160K context$0.196 in$0.28$0.28 out$0.4
DeepSeek
deepseek-v4-flash-30%Text

DeepSeek model with remarkable quality-per-dollar for chat and RAG.

by DeepSeekMay 20261M context$0.098 in$0.14$0.196 out$0.28
DeepSeek
deepseek-v4-pro-30%Reasoning

DeepSeek model with remarkable quality-per-dollar for chat and RAG.

by DeepSeekMay 20261M context$0.3045 in$0.435$0.609 out$0.87
Grok
grok-4.5-30%Reasoning

xAI Grok reasoning model for coding, analysis and tool-driven agent workflows.

by Grok— context$1.4 in$2$4.2 out$6
Grok
grok-4.6-30%Reasoning

xAI Grok reasoning model for coding, analysis and tool-driven agent workflows.

by Grok— context$1.4 in$2$4.2 out$6
Kimi
kimi-k2-0905-preview-30%Text

Moonshot Kimi open-weights MoE model with strong agentic and multilingual performance.

by Moonshot KimiSep 2025262K context$0.42 in$0.6$1.75 out$2.5
Kimi
kimi-k2-thinking-30%Reasoning

Moonshot Kimi reasoning model interleaving step-by-step thinking with tool calls.

by Moonshot KimiNov 2025256K context$0.42 in$0.6$1.75 out$2.5
Kimi
kimi-k2-thinking-turbo-30%Reasoning

Moonshot Kimi reasoning model interleaving step-by-step thinking with tool calls.

by Moonshot KimiNov 2025262K context$0.805 in$1.15$5.6 out$8
Kimi
kimi-k2-turbo-preview-30%Text

High-throughput turbo tier of Moonshot Kimi for latency-sensitive workloads.

by Moonshot KimiSep 2025262K context$0.805 in$1.15$5.6 out$8
Kimi
kimi-k2.5-30%Text

Moonshot’s open-weights MoE flagship with strong agentic and multilingual performance.

by Moonshot KimiJan 2026262K context$0.42 in$0.6$2.1 out$3
Kimi
kimi-k2.6-30%Text

Moonshot Kimi open-weights MoE model with strong agentic and multilingual performance.

by Moonshot KimiFeb 2026262K context$0.665 in$0.95$2.8 out$4
Kimi
kimi-k2.7-code-30%Coding

Moonshot Kimi coding model tuned for agentic software engineering.

by Moonshot KimiApr 2026262K context$0.665 in$0.95$2.8 out$4
Kimi
kimi-k3-30%Reasoning

Moonshot’s third-generation flagship MoE — frontier-level agentic and reasoning performance at aggressive open pricing.

by Moonshot KimiJun 2026512K context$0.7 in$1$2.8 out$4
Kimi
kimi-latest-30%Text

Moonshot Kimi open-weights MoE model with strong agentic and multilingual performance.

by Moonshot Kimi128K context$1.4 in$2$3.5 out$5
OpenAI
gpt-5-30%Text

General-purpose GPT model with strong reasoning and structured output.

by OpenAIAug 2025272K context$0.875 in$1.25$7 out$10
OpenAI
gpt-5-codex-30%Coding

Codex variant tuned for agentic software engineering and long coding sessions.

by OpenAISep 2025272K context$0.875 in$1.25$7 out$10
OpenAI
gpt-5-mini-30%Text

Cost-efficient GPT tier for everyday chat, RAG and high-throughput workloads.

by OpenAIAug 2025272K context$0.175 in$0.25$1.4 out$2
OpenAI
gpt-5-nano-30%Text

Cost-efficient GPT tier for everyday chat, RAG and high-throughput workloads.

by OpenAIAug 2025272K context$0.035 in$0.05$0.28 out$0.4
OpenAI
gpt-5-pro-30%Reasoning

Pro tier of the GPT line for maximum quality on the hardest problems.

by OpenAIOct 2025400K context$10.5 in$15$84 out$120
OpenAI
gpt-5.1-30%Text

General-purpose GPT model with strong reasoning and structured output.

by OpenAINov 2025272K context$0.875 in$1.25$7 out$10
OpenAI
gpt-5.1-codex-30%Coding

Codex variant tuned for agentic software engineering and long coding sessions.

by OpenAINov 2025272K context$0.875 in$1.25$7 out$10
OpenAI
gpt-5.1-codex-mini-30%Coding

Codex variant tuned for agentic software engineering and long coding sessions.

by OpenAINov 2025272K context$0.175 in$0.25$1.4 out$2
OpenAI
gpt-5.2-30%Reasoning

OpenAI’s flagship general-purpose model with strong reasoning, multimodal understanding and reliable structured output.

by OpenAIDec 2025272K context$1.225 in$1.75$9.8 out$14
OpenAI
gpt-5.2-codex-30%Coding

Agentic coding variant of GPT-5.2 tuned for long-running software engineering sessions, refactors and code review.

by OpenAIDec 2025272K context$1.225 in$1.75$9.8 out$14
OpenAI
gpt-5.2-mini-30%Text

Cost-efficient GPT tier for everyday chat, RAG and high-throughput workloads.

by OpenAIDec 2025272K context$0.175 in$0.25$1.4 out$2
OpenAI
gpt-5.2-pro-30%Reasoning

Pro tier of the GPT line for maximum quality on the hardest problems.

by OpenAIDec 2025272K context$14.7 in$21$117.6 out$168
OpenAI
gpt-5.3-codex-30%Coding

Codex variant tuned for agentic software engineering and long coding sessions.

by OpenAIFeb 2026272K context$1.225 in$1.75$9.8 out$14
OpenAI
gpt-5.4-30%Reasoning

General-purpose GPT model with strong reasoning and structured output.

by OpenAIMar 20261M context$1.75 in$2.5$10.5 out$15
OpenAI
gpt-5.4-mini-30%Text

Cost-efficient GPT tier for everyday chat, RAG and high-throughput workloads.

by OpenAIMar 20261M context$0.525 in$0.75$3.15 out$4.5
OpenAI
gpt-5.4-nano-30%Text

Cost-efficient GPT tier for everyday chat, RAG and high-throughput workloads.

by OpenAIMar 20261M context$0.14 in$0.2$0.875 out$1.25
OpenAI
gpt-5.4-pro-30%Reasoning

Pro tier of the GPT line for maximum quality on the hardest problems.

by OpenAIMar 20261M context$21 in$30$126 out$180
OpenAI
gpt-5.5-30%Reasoning

General-purpose GPT model with strong reasoning and structured output.

by OpenAIApr 20261M context$3.5 in$5$21 out$30
OpenAI
gpt-5.5-pro-30%Reasoning

Pro tier of the GPT line for maximum quality on the hardest problems.

by OpenAIApr 20261M context$21 in$30$126 out$180
OpenAI
gpt-5.6-30%Reasoning

Latest GPT generation with a 1M-token window; the strongest general model in the GPT line-up.

by OpenAIJun 20261M context$3.5 in$5$21 out$30
OpenAI
gpt-5.6-luna-30%Text

Cost-efficient GPT tier for everyday chat, RAG and high-throughput workloads.

by OpenAIJun 20261M context$0.7 in$1$4.2 out$6
OpenAI
gpt-5.6-sol-30%Reasoning

General-purpose GPT model with strong reasoning and structured output.

by OpenAIJun 20261M context$3.5 in$5$21 out$30
OpenAI
gpt-5.6-terra-30%Text

General-purpose GPT model with strong reasoning and structured output.

by OpenAIJun 20261M context$1.75 in$2.5$10.5 out$15
OpenAI
gpt-image-2-30%Image

OpenAI image-generation model for high-quality text-to-image and image editing, priced per generated image by output resolution.

by OpenAIApr 2026— context1K $0.03 / image$0.04292K $0.05 / image$0.07144K $0.06 / image$0.0857
ChatGLM
glm-4.5-30%Text

Zhipu GLM model with strong coding and agent capabilities.

by Z.AI GLMJul 2025128K context$0.42 in$0.6$1.54 out$2.2
ChatGLM
glm-4.5-air-30%Text

Lightweight GLM tier for fast, cheap, high-volume completions.

by Z.AI GLMJul 2025128K context$0.14 in$0.2$0.77 out$1.1
ChatGLM
glm-4.6-30%Coding

Zhipu’s workhorse GLM model — a popular Claude-compatible daily driver for coding and agents.

by Z.AI GLMSep 2025200K context$0.42 in$0.6$1.54 out$2.2
ChatGLM
glm-4.7-30%Coding

Zhipu GLM model with strong coding and agent capabilities.

by Z.AI GLMDec 2025200K context$0.42 in$0.6$1.54 out$2.2
ChatGLM
glm-4.7-flash-30%Text

Lightweight GLM tier for fast, cheap, high-volume completions.

by Z.AI GLMDec 2025200K context$0.049 in$0.07$0.28 out$0.4
ChatGLM
glm-5-30%Text

Zhipu’s fifth-generation flagship with strong coding and agent capabilities.

by Z.AI GLMApr 2026200K context$0.7 in$1$2.24 out$3.2
ChatGLM
glm-5-code-30%Coding

Zhipu GLM coding model tuned for software engineering and agent use.

by Z.AI GLMApr 2026200K context$0.84 in$1.2$3.5 out$5
ChatGLM
glm-5.1-30%Reasoning

Zhipu GLM model with strong coding and agent capabilities.

by Z.AI GLMJun 2026200K context$0.98 in$1.4$3.08 out$4.4
ChatGLM
glm-5.2-30%Reasoning

Zhipu GLM model with strong coding and agent capabilities.

by Z.AI GLM200K context$0.7 in$1$2.24 out$3.2
ChatGLM
glm-5.3-30%Reasoning

Zhipu GLM model with strong coding and agent capabilities.

by Z.AI GLM200K context$0.7 in$1$2.24 out$3.2