Models

anthropic

Claude Fable 5.1

×8

An update to Claude Fable 5 from Anthropic: the top price tier . Improved reasoning, coding and long agentic chains.

deepseek

DeepSeek V4.1 Flash

×0.3

An update to DeepSeek's cost-efficient line: V4.1 Flash is faster and more accurate than V4 Flash while keeping the same long context up to 1M tokens. A fit for high-volume agentic workloads and fast assistants.

openai

GPT-6 Astra

×7.5

OpenAI's new flagship: GPT-6 Astra. The strongest model in the lineup for complex reasoning, coding and agentic workloads. 1M-token context.

openai

GPT-6 Sol

×2

The mid tier of the GPT-6 family between flagship Astra and budget Luna: complex reasoning, coding and agentic tasks at a noticeably lower price than Astra. 1M-token context.

openai

GPT-6 Luna

×0.25

The fastest and cheapest tier of the GPT-6 family for high-throughput workloads: drafts, classification, bulk calls. 1M-token context.

google

Gemini 3.8 Flash

×1

Google's newest fast model with separate low, medium and high reasoning modes, multimodal input and a 1M-token context.

z-ai

GLM 5.3

×1.9

A large Z.ai reasoning model for complex engineering and long agentic runs: 1M-token context, up to 131K output. Reasoning is always on (low / high / max, max by default), with better token economy than GLM-5.2.

z-ai

GLM 5.3 Flash

×0.6

A fast, low-cost version of GLM 5.3: the same 1M-token context and up to 131K output, but significantly cheaper and with lower latency. Built for high-throughput agentic runs, code completion and bulk text processing.

google

Gemini 3.7 Flash

×1

Google's fast Antigravity model with separate low, medium and high reasoning modes, multimodal input and a 1M-token context.

x-ai

Grok 4.6

×0.8

xAI's strongest model: frontier results in code, knowledge work and STEM. Accepts text, images and files (including PDF), 500K-token context. Widely used in agentic developer tooling.

deepseek

DeepSeek V4 Pro

×0.6

DeepSeek's large MoE model, the general-availability V4 Pro release. 1M-token context, served both by DeepSeek and third-party inference providers; strong results in reasoning, code and agentic benchmarks.

nvidia

Nemotron 3.5 Lightning

×0.05

An open lightweight MoE model: 3B active parameters out of 30B total, built for high throughput in agentic workloads and domain fine-tuning. 1M-token context, part of NVIDIA's open model family for agentic AI.

qwen

Qwen3.8 Max

×0.3

The flagship of Alibaba's Qwen3.8 series: a multimodal reasoning model for hard tasks, visual understanding, code and agentic work. 1M-token context and top placements in the Artificial Analysis intelligence, coding and agentic indexes.

moonshotai

Kimi K3

×5.5

Moonshot AI's open natively multimodal model with 2.8T parameters — the first open model in the 3T class. Built on Kimi Delta Attention and Attention Residuals with a 1M-token context; designed for long-horizon code, knowledge work and reasoning.

openai

GPT-5.6 Sol

×3

The flagship of the GPT-5.6 series: complex reasoning, coding and agentic workloads. Especially strong on multi-step terminal tasks and long chains of code work. 1M-token context, world knowledge up to February 2026.

openai

GPT-5.6 Terra

×1.5

The balanced GPT-5.6 tier between the flagship Sol and the budget Luna. Built for everyday coding, reasoning and agentic tasks at roughly half the price of Sol. 1M-token context.

openai

GPT-5.6 Luna

×0.22

The fastest and cheapest tier of the GPT-5.6 family for high-throughput workloads. Supports reasoning-effort control and prompt caching, just like the larger models in the series.

x-ai

Grok 4.5

×0.8

A frontier xAI model with strong results in code, knowledge work and STEM. 500K-token context with step-by-step reasoning. Widely used in coding assistants and autonomous agent frameworks.

google

Gemini 3.6 Flash

×1

Google's new fast Antigravity model with separate low, medium and high reasoning modes, multimodal input and a 1M-token context.

anthropic

Claude Sonnet 5

×2

The most agentic Sonnet model: it closes the gap to Opus 4.8 in reasoning, tool use and code at a markedly lower price. Hallucinates and sycophantically agrees less often than Sonnet 4.6.

moonshotai

Kimi K2.7 Code

×1.5

The coding model of the Kimi K2 family, based on K2.6: a 1T-parameter MoE with 32B active (384 experts, 8 per token) and the MoonViT vision encoder. 262K context, always in reasoning mode with the chain preserved between turns.

nvidia

Nemotron 3 Ultra

×0.05

An open frontier model for reasoning and orchestration: 550B parameters, 55B active, a hybrid Transformer-Mamba MoE architecture with LatentMoE and native speculative decoding via MTP. Up to 1M context, tuned for long agentic workflows.

qwen

Qwen3.7 Plus

×0.1

The cost-efficient Qwen3.7 model with text and image input. Vision is substantially stronger while agentic intelligence is preserved: it reads scenes and interfaces, generates code from visual references and can navigate mobile apps.

minimax

MiniMax M3

×0.45

MiniMax's multimodal model: text, image and video input with a 1M-token context. Built on MiniMax Sparse Attention with KV-block selection instead of full attention — roughly 20× less compute per token at long context.

anthropic

Claude Opus 4.8

×4

An evolution of Opus 4.7: better code, more reliable agentic runs, roughly 4× less likely to let faulty code pass without comment. Adds user-set effort levels and Claude Code "dynamic workflows" with parallel subagents.

qwen

Qwen3.7 Max

×0.2

The Qwen3.7 flagship: text in, text out, 1M-token context. Aimed at agentic workloads — coding, office tasks and long autonomous execution — with prompt caching support.

cursor

Composer 2.5 (Fast)

×0.5

Cursor's own agentic coding model, trained from a Moonshot Kimi K2.5 checkpoint with extra RL on long-horizon tasks. The Fast variant is the default interactive mode, tuned for tool use, multi-file edits and the terminal.

google

Gemini 3.5 Flash

×1

Google's fast, solid-quality model: good accuracy, high speed and a 1M-token context. Simpler than the Pro tier, but it answers faster and costs less.

openai

GPT-5.5

×3

OpenAI's spring 2026 flagship: the most agentic model in the lineup at launch, 82.7% on Terminal-Bench 2.0, strong at computer use and knowledge work. Same speed per token as GPT-5.4, but it spends fewer tokens on equivalent Codex tasks.

deepseek

DeepSeek V4 Flash

×0.1

DeepSeek's cost-efficient MoE model: 284B parameters, 13B active, hybrid attention for cheap long-context work up to 1M tokens. Supports high and xhigh reasoning effort — a fit for fast assistants and high-volume agents.

xiaomi

MiMo-V2.5-Pro

×0.3

Xiaomi's flagship trillion-parameter agentic model: strong results on ClawEval, GDPVal and SWE-bench Pro. Sustains coherent autonomous execution across nearly 1,000 tool calls, with up to 1M context and 128K output.

xiaomi

MiMo-V2.5

×0.05

Xiaomi's natively omni-modal model: Pro-level agentic performance at roughly half the inference cost, and better image and video understanding than MiMo-V2-Omni. The 1M-token context fits whole documents and long conversations.

openai

GPT-5.4 Mini

×0.8

A fast, cost-efficient version of GPT-5.4 for high throughput. Handles text and images, works confidently with tools — a fit for chat apps, coding assistants and high-volume agentic workloads.

z-ai

GLM 5 Turbo

×1.38

A Z.ai model for fast inference in agentic environments with long execution chains. 203K context, with improved decomposition of complex instructions, tool use and stability across long runs.

google

Gemini 3.1 Pro

×2

Google's strong high-end general-purpose model: high accuracy, deep reasoning and a 1M-token context. Accepts text, images, audio, video and PDF, and replies with text.

openai

GPT-5.4

×1.5

The model that merged the Codex and GPT lines into one system: 922K input context, up to 128K output, text and image input. Improved coding, document understanding, tool use and instruction following.

anthropic

Claude Opus 4.6

×4

The first Opus-class model with a 1M-token context (beta): plans more carefully and sustains agentic tasks longer across large codebases. SOTA on Terminal-Bench 2.0 and Humanity's Last Exam at release.

openai

GPT-4o Transcribe

×1

OpenAI's speech recognition model: it takes an audio file and returns text via the /v1/audio/transcriptions route. A fit for transcribing calls, interviews and voice messages.

openai

GPT-6.1 Sol

×3

GPT-6.1 Sol. Model multiplier: ×3.

anthropic

Claude Fable 5

×8

Anthropic's top price tier — the same rate as the limited-availability Claude Mythos 5. Shipped with stricter cybersecurity safeguards than the later Sonnet models.

anthropic

Claude Opus 5.5

×4

The latest refresh of the Opus lineup at the same rate as Opus 5 . Built for long-running agentic coding and knowledge work: holds multi-step tasks more reliably and works longer without losing focus.

anthropic

Claude Opus 5

×4

The Opus-class flagship: — the same rate as Opus 4.5–4.8. Claude Sonnet 5 is reported to approach Opus 5, which makes the latter the accuracy benchmark of the lineup.

anthropic

Claude Opus 4.7

×4

An Opus-class model, the predecessor of Opus 4.8, vendor price . Used as the comparison baseline in benchmarks for code, computer use and legal agents.

anthropic

Claude Sonnet 4.6

×2

The predecessor of Claude Sonnet 5, vendor price . Served as the baseline against which Sonnet 5 showed gains in agentic reasoning, tool use, code and safety.

anthropic

Claude Haiku 4.5

×0.9

Anthropic's fast, low-cost model, vendor price . Part of the Claude 4.5 generation that introduced regional and multi-region endpoints on Bedrock and Google Cloud.

z-ai

GLM 5.2

×1.9

A Z.ai reasoning model with a 1M-token context for long agentic scenarios, project-scale development and multi-step automation. Supports high and xhigh reasoning effort (xhigh being the maximum).

moonshotai

Kimi K2.8 Preview

×0.9

Kimi K2.8 Preview by Moonshot AI. Model multiplier: ×0.9.

moonshotai

Kimi K3.1 Pro

×2.5

Kimi K3.1 Pro by Moonshot AI. Model multiplier: ×2.5.

xiaomi

MiMo-V2.6-Pro

×0.3

MiMo-V2.6-Pro by Xiaomi. Model multiplier: ×0.3.

xiaomi

MiMo-V2.6-Flash

×0.1

MiMo-V2.6-Flash by Xiaomi. Model multiplier: ×0.1.

minimax

MiniMax M3.1 Flash

×0.3

MiniMax M3.1 Flash. Model multiplier: ×0.3.

A light, very cheap model for high-volume work: classification, data extraction, short answers and throughput-heavy scenarios with a minimal cost per token.