Skip to main content
Home / Models / LLM

Large Language Models

Every LLM we serve โ€” from Claude and Gemini to DeepSeek and Qwen. Compare context windows, pricing, reasoning capabilities, and real-world performance. No benchmark hype, just practical data from someone who's integrated them all.

Live data from 51 providers โ€” updated daily
Showing 141 of 141 LLMs
ByteDance Seed 1.8
BytePlus
ByteDance Seed 1.8, multimodal agent model (text, image & video in โ†’ text), strong tool use + reasoning, 256K context. Native via BytePlus (Ark).
262K
66K outIn: $0.000438/1MOut: $0.0035/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimagevideo
Try on Zubnet โ†’
ByteDance Seed 2.0 Code
BytePlus
ByteDance Seed 2.0 Code (preview), coding-specialized agent (text & image in โ†’ text), strong tool use + reasoning. Native via BytePlus (Ark).
262K
66K outIn: $0.000875/1MOut: $0.0053/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
ByteDance Seed 2.0 Lite
BytePlus
ByteDance Seed 2.0 Lite, efficient multimodal agent (text, image & video in โ†’ text), tool use + reasoning, 256K context. Native via BytePlus (Ark).
262K
66K outIn: $0.000438/1MOut: $0.0035/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsreasoningimagevideo
Try on Zubnet โ†’
ByteDance Seed 2.0 Mini
BytePlus
ByteDance Seed 2.0 Mini, fast lightweight multimodal agent (text, image & video in โ†’ text), tool use. Native via BytePlus (Ark).
131K
33K outIn: $0.000175/1MOut: $0.000700/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoningimagevideo
Try on Zubnet โ†’
ByteDance Seed 2.0 Pro
BytePlus
ByteDance Seed 2.0 Pro, flagship multimodal agent (text, image & video in โ†’ text), strong tool use + reasoning, 256K context. Native via BytePlus (Ark).
262K
66K outIn: $0.000875/1MOut: $0.0053/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimagevideo
Try on Zubnet โ†’
ByteDance Seed 2.1 Turbo
BytePlus
ByteDance Seed 2.1 Turbo, fast multimodal reasoning agent (text & image in โ†’ text), strong tool use + thinking, 256K context. Native via BytePlus (Ark).
262K
66K outIn: $0.000875/1MOut: $0.0044/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
Codestral
Mistral AI
Specialized model for code generation, completion, and understanding.
262K
8K outIn: $0.000525/1MOut: $0.0016/1MSpeed: โ—โ—โ—โ—โ—‹
Streamchat
Try on Zubnet โ†’
DeepSeek V4 Pro
DeepSeek
DeepSeek flagship V4. 512K context, advanced reasoning + tool use.
524K
66K outIn: $0.0012/1MOut: $0.0035/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamchatreasoningtools
Try on Zubnet โ†’
DeepSeek V4.1 Flash
DeepSeek
New
DeepSeek V4.1 Flash, the fast daily driver that replaced V4 Flash on 2026-09-10. 1M context, thinking (default) and non-thinking modes.
1M
393K outIn: $0.000262/1MOut: $0.0010/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamchatreasoningtools
Try on Zubnet โ†’
Fable 5
Anthropic
Fable 5 by Anthropic.
1M
128K outIn: $0.0175/1MOut: $0.0875/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
Fable 5.1
Anthropic
New
Fable 5.1 by Anthropic: the most capable generally available Claude, for demanding reasoning and long-horizon agentic work. 1M context, 128K output, adaptive thinking always on.
1M
128K outIn: $0.0175/1MOut: $0.0875/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
Fugu Ultra
Sakana
Sakana's multi-agent conductor, orchestrates a pool of frontier models for complex reasoning. Billed per token, orchestration included.
272K
16K outIn: $0.0088/1MOut: $0.0525/1MSpeed: โ—โ—โ—‹โ—‹โ—‹
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
Gemini 2.5 Flash
Google
Fast and efficient model with adaptive thinking for complex tasks
1M
66K outIn: $0.000525/1MOut: $0.0044/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoningimagevideo
Try on Zubnet โ†’
Gemini 2.5 Flash-Lite
Google
Cost-efficient high-throughput model for budget-conscious applications
1M
66K outIn: $0.000175/1MOut: $0.000700/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoningimagevideo
Try on Zubnet โ†’
Gemini 2.5 Pro
Google
Enhanced thinking and reasoning for complex problems
1M
66K outIn: $0.0022/1MOut: $0.0175/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimagevideo
Try on Zubnet โ†’
Gemini 3 Flash
Google
Google's fastest frontier model. Beats 2.5 Pro at 1/4 the cost with 1M context
1M
66K outIn: $0.000875/1MOut: $0.0053/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoningimagevideo
Try on Zubnet โ†’
Gemini 3.1 Flash-Lite
Google
Cost-efficient high-throughput model for budget-conscious applications.
1M
66K outIn: $0.000438/1MOut: $0.0026/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
Gemini 3.1 Pro
Google
Google's most capable agentic model with 1M token context, 77.1% ARC-AGI-2 reasoning, and native tool use.
1M
66K outIn: $0.0035/1MOut: $0.0210/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimagevideo+1
Try on Zubnet โ†’
Gemini 3.5 Flash
Google
Fast flagship with adaptive thinking for complex tasks.
1M
66K outIn: $0.0026/1MOut: $0.0158/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoningimagevideo
Try on Zubnet โ†’
Gemini 3.5 Flash-Lite
Google
Cost-efficient GA Flash model for high-throughput multimodal tasks.
1M
66K outIn: $0.000525/1MOut: $0.0044/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoningimagevideo
Try on Zubnet โ†’
Gemini 3.6 Flash
Google
Google's latest GA Flash model for fast multimodal reasoning and agentic workloads.
1M
66K outIn: $0.0026/1MOut: $0.0131/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoningimagevideo
Try on Zubnet โ†’
Gemini 3.7 Flash
Google
Google's most capable Flash model for agentic workflows and multimodal reasoning.
1.048576M
66K outIn: $0.0026/1MOut: $0.0131/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoningimagevideo
Try on Zubnet โ†’
Gemma 4 26B MoE
Google
Google's open-weight mixture-of-experts model, 26B total, 4B active parameters for fast inference
262K
33K outIn: $0.000105/1MOut: $0.000577/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
Gemma 4 31B
Google
Google's open-weight dense model with 262K context and strong multilingual reasoning
262K
33K outIn: $0.000245/1MOut: $0.000700/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
GLM-4.5
Z.ai
Previous flagship with thinking mode. MoE architecture, 128K context.
128K
8K outIn: $0.0010/1MOut: $0.0039/1MSpeed: โ—โ—โ—โ—โ—‹
Streamreasoning
Try on Zubnet โ†’
GLM-4.5 Air
Z.ai
Lightweight model optimized for efficiency. 128K context.
128K
8K outIn: $0.000350/1MOut: $0.0019/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
GLM-4.5 Flash
Z.ai
Free tier model. Great for testing and light workloads.
128K
4K outSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
GLM-4.5V
Z.ai
Vision-language model for image understanding and analysis.
66K
16K outIn: $0.0010/1MOut: $0.0032/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
GLM-4.6
Z.ai
Flagship model with reasoning, coding, and agentic capabilities. 128K context.
128K
8K outIn: $0.0010/1MOut: $0.0039/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
GLM-4.6V
Z.ai
Vision-capable model for image understanding and analysis.
128K
8K outIn: $0.000525/1MOut: $0.0016/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
GLM-4.6V Flash
Z.ai
Free vision model. Great for testing image understanding.
128K
4K outSpeed: โ—โ—โ—โ—โ—
Streamimage
Try on Zubnet โ†’
GLM-4.6V FlashX
Z.ai
Lightweight vision model. Fast and cost-effective for image tasks.
128K
4K outIn: $0.000070/1MOut: $0.000700/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamreasoningtoolsimage
Try on Zubnet โ†’
GLM-4.7
Z.ai
Latest flagship. 358B params, 204K context, 131K output. #1 on LiveCodeBench.
205K
131K outIn: $0.0010/1MOut: $0.0039/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
GLM-4.7 Flash
Z.ai
Lightweight, completely free GLM-4.7 variant with 200K context.
205K
131K outSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
GLM-4.7 FlashX
Z.ai
Lightweight, high-speed and affordable GLM-4.7 variant with 200K context.
205K
131K outIn: $0.000122/1MOut: $0.000700/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
GLM-5
Z.ai
Most capable Z.ai model. 744B params (40B active MoE), 28.5T training tokens. Built for complex systems engineering and agentic tasks.
205K
131K outIn: $0.0018/1MOut: $0.0056/1MSpeed: โ—โ—โ—‹โ—‹โ—‹
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
GLM-5 Turbo
Z.ai
Fast variant of GLM-5 with optimized speed and competitive quality.
In: $0.0021/1MOut: $0.0070/1M
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
GLM-5.1
Z.ai
Z.ai's current flagship for agentic and coding tasks.
205K
131K outIn: $0.0024/1MOut: $0.0077/1MSpeed: โ—โ—โ—‹โ—‹โ—‹
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
GLM-5.2
Z.ai
Z.ai's latest flagship for agentic and coding tasks.
205K
131K outIn: $0.0024/1MOut: $0.0077/1MSpeed: โ—โ—โ—‹โ—‹โ—‹
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
GLM-5.3
Z.ai
Z.ai's newest flagship for agentic and coding tasks. Always reasons; text-only input.
205K
131K outIn: $0.0024/1MOut: $0.0077/1MSpeed: โ—โ—โ—‹โ—‹โ—‹
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
GLM-5.3 Flash
Z.ai
New
Z.ai's fast, low-cost GLM-5.3 variant: 1M context, tools, always reasons; text and image input.
1M
131K outIn: $0.000262/1MOut: $0.000875/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsreasoningvisionimage
Try on Zubnet โ†’
GLM-5V Turbo
Z.ai
Native multimodal coding model. 744B MoE (40B active), 203K context. Optimized for design-to-code, GUI automation, and vision-grounded agentic tasks.
205K
131K outIn: $0.0021/1MOut: $0.0070/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsreasoningimagevideo
Try on Zubnet โ†’
GPT-3.5 Turbo
OpenAI
OpenAI GPT-3.5 Turbo (chat completions endpoint).
16K
4K outIn: $0.000875/1MOut: $0.0026/1M
tools
Try on Zubnet โ†’
GPT-3.5 Turbo 16K
OpenAI
OpenAI GPT-3.5 Turbo 16K (chat completions endpoint).
16K
4K outIn: $0.0053/1MOut: $0.0070/1M
tools
Try on Zubnet โ†’
GPT-4
OpenAI
OpenAI GPT-4 (chat completions endpoint).
8K
8K outIn: $0.0525/1MOut: $0.1050/1M
tools
Try on Zubnet โ†’
GPT-4 Turbo
OpenAI
OpenAI GPT-4 Turbo (chat completions endpoint).
128K
4K outIn: $0.0175/1MOut: $0.0525/1M
toolsvisionimage
Try on Zubnet โ†’
GPT-4.1
OpenAI
OpenAI GPT-4.1 (chat completions endpoint).
1.047576M
33K outIn: $0.0035/1MOut: $0.0140/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsimage
Try on Zubnet โ†’
GPT-4.1 Mini
OpenAI
OpenAI GPT-4.1 Mini (chat completions endpoint).
1.047576M
33K outIn: $0.000700/1MOut: $0.0028/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsimage
Try on Zubnet โ†’
GPT-4o
OpenAI
OpenAI GPT-4o (chat completions endpoint).
128K
16K outIn: $0.0044/1MOut: $0.0175/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsimage
Try on Zubnet โ†’
GPT-4o Mini
OpenAI
OpenAI GPT-4o Mini (chat completions endpoint).
128K
16K outIn: $0.000262/1MOut: $0.0010/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsimage
Try on Zubnet โ†’
GPT-5
OpenAI
OpenAI GPT-5 (chat completions endpoint).
272K
128K outIn: $0.0022/1MOut: $0.0175/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsimage
Try on Zubnet โ†’
GPT-5 Mini
OpenAI
OpenAI GPT-5 Mini (chat completions endpoint).
272K
128K outIn: $0.000438/1MOut: $0.0035/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsimage
Try on Zubnet โ†’
GPT-5 Nano
OpenAI
OpenAI GPT-5 Nano (chat completions endpoint).
272K
128K outIn: $0.000087/1MOut: $0.000700/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsimage
Try on Zubnet โ†’
GPT-5 Pro
OpenAI
OpenAI GPT-5 Pro (chat completions endpoint).
272K
128K outIn: $0.0262/1MOut: $0.2100/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
GPT-5 Search API
OpenAI
OpenAI GPT-5 Search API (chat completions endpoint).
272K
128K outIn: $0.0022/1MOut: $0.0175/1MSpeed: โ—โ—โ—โ—โ—‹
Streamweb_searchimage
Try on Zubnet โ†’
GPT-5.1
OpenAI
OpenAI GPT-5.1 (chat completions endpoint).
272K
128K outIn: $0.0022/1MOut: $0.0175/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsimage
Try on Zubnet โ†’
GPT-5.2
OpenAI
OpenAI GPT-5.2 (chat completions endpoint).
272K
128K outIn: $0.0031/1MOut: $0.0245/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsimage
Try on Zubnet โ†’
GPT-5.2 Pro
OpenAI
OpenAI GPT-5.2 Pro (chat completions endpoint).
272K
128K outIn: $0.0367/1MOut: $0.2940/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
GPT-5.3 Codex
OpenAI
OpenAI GPT-5.3 Codex (chat completions endpoint).
272K
128K outIn: $0.0031/1MOut: $0.0245/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
GPT-5.4
OpenAI
OpenAI mid-tier flagship (March 2026).
272K
128K outIn: $0.0044/1MOut: $0.0262/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsimage
Try on Zubnet โ†’
GPT-5.4 Mini
OpenAI
OpenAI fast/cheap chat model.
272K
128K outIn: $0.0013/1MOut: $0.0079/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsimage
Try on Zubnet โ†’
GPT-5.4 Nano
OpenAI
OpenAI cheapest tier, very fast.
272K
128K outIn: $0.000350/1MOut: $0.0022/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsimage
Try on Zubnet โ†’
GPT-5.4 Pro
OpenAI
OpenAI GPT-5.4 Pro (chat completions endpoint).
272K
128K outIn: $0.0525/1MOut: $0.3150/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
GPT-5.5
OpenAI
OpenAI flagship chat model (April 2026 release).
272K
128K outIn: $0.0088/1MOut: $0.0525/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsimage
Try on Zubnet โ†’
GPT-5.5 Pro
OpenAI
OpenAI premium tier with deep reasoning. April 2026.
272K
128K outIn: $0.0525/1MOut: $0.3150/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
GPT-5.6 Luna
OpenAI
OpenAI GPT-5.6 Luna โ€” high-volume streaming tasks, fast + economical 5.6 tier.
272K
128K outIn: $0.0018/1MOut: $0.0105/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoningvisionimage
Try on Zubnet โ†’
GPT-5.6 Sol
OpenAI
OpenAI GPT-5.6 Sol โ€” frontier logic + deep reasoning, the flagship 5.6 tier.
272K
128K outIn: $0.0088/1MOut: $0.0525/1MSpeed: โ—โ—โ—‹โ—‹โ—‹
ToolsStreamtoolsreasoningvisionimage
Try on Zubnet โ†’
GPT-5.6 Terra
OpenAI
OpenAI GPT-5.6 Terra โ€” balanced business logic, mid 5.6 tier.
272K
128K outIn: $0.0044/1MOut: $0.0262/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningvisionimage
Try on Zubnet โ†’
GPT-6 Astra
OpenAI
New
OpenAI GPT-6 Astra โ€” frontier reasoning over a 1M-token context window.
1.05M
128K outIn: $0.0175/1MOut: $0.0875/1MSpeed: โ—โ—โ—‹โ—‹โ—‹
ToolsStreamtoolsreasoningvisionimage
Try on Zubnet โ†’
GPT-6 Luna
OpenAI
New
OpenAI GPT-6 Luna: the fast, low-cost GPT-6 tier for high-volume chat and light agentic work, 1M-token context window.
1.05M
128K outIn: $0.000175/1MOut: $0.000875/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoningvisionimage
Try on Zubnet โ†’
GPT-6 Sol
OpenAI
New
OpenAI GPT-6 Sol: GPT-6 reasoning at a fifth of Astra's price, over a 1M-token context window.
1.05M
128K outIn: $0.0035/1MOut: $0.0175/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningvisionimage
Try on Zubnet โ†’
Grok 4.20 Fast
xAI / Grok
Non-reasoning variant of Grok 4.20. Fastest 4.20 path for chat and high-throughput tasks.
1M
8K outIn: $0.0022/1MOut: $0.0044/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamchattoolsimage
Try on Zubnet โ†’
Grok 4.20 Reasoning
xAI / Grok
Reasoning-mode variant of Grok 4.20. Slower but stronger on multi-step problems.
1M
8K outIn: $0.0022/1MOut: $0.0044/1MSpeed: โ—โ—โ—‹โ—‹โ—‹
ToolsStreamchatreasoningtoolsimage
Try on Zubnet โ†’
Grok 4.3
xAI / Grok
xAI's flagship general-purpose model. 1M context, balanced reasoning and chat.
1M
8K outIn: $0.0022/1MOut: $0.0044/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamchattoolsreasoningimage
Try on Zubnet โ†’
Grok 4.5
xAI / Grok
xAI's newest flagship, launched July 2026. 500K context, strongest reasoning + coding, trained alongside Cursor.
500K
8K outIn: $0.0035/1MOut: $0.0105/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamchattoolsreasoningimage
Try on Zubnet โ†’
Grok 4.6
xAI / Grok
xAI's newest flagship. 500K context, accepts text and images, strongest reasoning + coding.
500K
8K outIn: $0.0035/1MOut: $0.0105/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamchattoolsreasoningimage
Try on Zubnet โ†’
Grok 4.7
xAI / Grok
New
xAI's newest flagship (2026-09-21). 500K context, accepts text and images, function calling, structured outputs, reasoning.
500K
8K outIn: $0.0035/1MOut: $0.0105/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamchattoolsreasoningimage
Try on Zubnet โ†’
Grok Build 0.1
xAI / Grok
Compact model tuned for code generation and structured output. 256K context.
256K
8K outIn: $0.0018/1MOut: $0.0035/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamchattoolsimage
Try on Zubnet โ†’
Haiku 4.5
Anthropic
Claude's fastest and most intelligent Haiku model
200K
64K outIn: $0.0018/1MOut: $0.0088/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
Kimi K2.6
Moonshot AI
Latest Kimi K2 reasoning model. Replaces K2 thinking variants (k2-thinking and k2-thinking-turbo) being retired 2026-05-25.
262K
131K outIn: $0.0017/1MOut: $0.0070/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamchatreasoningfunction_callingagentic
Try on Zubnet โ†’
Kimi K2.7 Code
Moonshot AI
Coding-specialized Kimi K2 model (k2.7). 256K context, agentic tool use and function calling, tuned for software engineering.
262K
131K outIn: $0.0017/1MOut: $0.0070/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamchatreasoningfunction_callingagentic
Try on Zubnet โ†’
Kimi K2.7 Code Highspeed
Moonshot AI
High-speed variant of Kimi K2.7 Code for low-latency agentic software engineering.
262K
131K outIn: $0.0017/1MOut: $0.0070/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamchatreasoningfunction_callingagentic
Try on Zubnet โ†’
Kimi K3
Moonshot AI
Moonshot's open-weight 2.8T-parameter multimodal reasoning model with a 1M-token context window.
1.048576M
131K outIn: $0.0053/1MOut: $0.0262/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamchatreasoningfunction_callingagentic+1
Try on Zubnet โ†’
M2
MiniMax
Agentic capabilities with advanced reasoning and thinking process visibility
In: $0.000525/1MOut: $0.0021/1M
ToolsStream
Try on Zubnet โ†’
M2 Stable
MiniMax
Optimized for high concurrency and commercial use with advanced reasoning
In: $0.000525/1MOut: $0.0021/1M
ToolsStream
Try on Zubnet โ†’
M2.5
MiniMax
Frontier reasoning model with SOTA coding, agentic tool use, and complex real-world task performance
In: $0.000525/1MOut: $0.0021/1M
ToolsStream
Try on Zubnet โ†’
M2.5 Highspeed
MiniMax
Fast variant of M2.5 optimized for speed at ~100 tokens per second
In: $0.0010/1MOut: $0.0042/1M
ToolsStream
Try on Zubnet โ†’
M2.7
MiniMax
Autonomous real-world productivity with agentic collaboration, live debugging, and professional document generation
In: $0.000525/1MOut: $0.0021/1M
ToolsStream
Try on Zubnet โ†’
M2.7 Highspeed
MiniMax
Fast variant of M2.7 for latency-sensitive work. 200K context, text only.
200K
128K outIn: $0.0010/1MOut: $0.0042/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtools
Try on Zubnet โ†’
M3
MiniMax
MiniMax flagship LLM, 1M context.
In: $0.000525/1MOut: $0.0021/1M
ToolsStream
Try on Zubnet โ†’
Magistral Small
Mistral AI
Efficient reasoning model for everyday logic tasks.
131K
40K outIn: $0.000875/1MOut: $0.0026/1MSpeed: โ—โ—โ—โ—โ—‹
Streamchatreasoning
Try on Zubnet โ†’
MiMo V2.6 Flash
Xiaomi
New
Xiaomi's fast, low-cost MiMo model (V2.6, 2026-09). 1M context, accepts text and images, function calling and structured output, reasons before it answers.
1.048576M
131K outIn: $0.000245/1MOut: $0.000490/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamchattoolsreasoningimage
Try on Zubnet โ†’
MiMo V2.6 Pro
Xiaomi
New
Xiaomi's flagship MiMo model (V2.6, 2026-09). 1M context, accepts text and images, function calling and structured output, reasons before it answers.
1.048576M
131K outIn: $0.000761/1MOut: $0.0015/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamchattoolsreasoningimage
Try on Zubnet โ†’
MiMo V2.6 Pro UltraSpeed
Xiaomi
New
MiMo V2.6 Pro on Xiaomi's high-speed tier, at a premium price. 1M context, accepts text and images, function calling and structured output.
1.048576M
131K outIn: $0.0076/1MOut: $0.0152/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamchattoolsreasoningimage
Try on Zubnet โ†’
Ministral 3 14B
Mistral AI
High-performance edge model with vision. Best Ministral for complex tasks.
131K
8K outIn: $0.000350/1MOut: $0.000350/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamchattoolsimage
Try on Zubnet โ†’
Ministral 3 3B
Mistral AI
Ultra-lightweight edge model with vision. 3B params, runs on phones/laptops.
131K
8K outIn: $0.000175/1MOut: $0.000175/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamchattoolsimage
Try on Zubnet โ†’
Ministral 3 8B
Mistral AI
Compact edge model with vision. 8B params for local deployment.
131K
8K outIn: $0.000262/1MOut: $0.000262/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamchattoolsimage
Try on Zubnet โ†’
Mistral Large 3
Mistral AI
Open-weight flagship. 675B MoE (41B active), multimodal with vision. Apache 2.0 licensed.
131K
8K outIn: $0.000875/1MOut: $0.0026/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamchattoolsimage
Try on Zubnet โ†’
Mistral Medium
Mistral AI
Premier frontier-class multimodal model. Great balance between Large and Small.
131K
8K outIn: $0.0026/1MOut: $0.0131/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamchattoolsreasoning
Try on Zubnet โ†’
Mistral Small
Mistral AI
Fast and efficient model for simple tasks. Great balance of performance and cost.
131K
8K outIn: $0.000175/1MOut: $0.000525/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamchattools
Try on Zubnet โ†’
Mistral Vibe CLI
Mistral AI
Mistral's model optimized for command-line and terminal-based coding workflows.
128K
33K outIn: $0.000175/1MOut: $0.000525/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtools
Try on Zubnet โ†’
o3
OpenAI
OpenAI o3 (chat completions endpoint).
200K
100K outIn: $0.0035/1MOut: $0.0140/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
Opus 4.5
Anthropic
Premium model combining maximum intelligence with practical performance
200K
64K outIn: $0.0088/1MOut: $0.0437/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
Opus 4.6
Anthropic
Most capable Claude model with enhanced reasoning and coding
1M
64K outIn: $0.0088/1MOut: $0.0437/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
Opus 4.7
Anthropic
Highly capable Claude Opus for complex reasoning and coding (1M context)
1M
128K outIn: $0.0088/1MOut: $0.0437/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
Opus 4.8
Anthropic
Anthropic's most capable model for complex reasoning and agentic coding (1M context)
1M
128K outIn: $0.0088/1MOut: $0.0437/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
Opus 5
Anthropic
Anthropic's flagship Opus โ€” deep reasoning, agentic coding and long-horizon work (1M context)
1M
128K outIn: $0.0088/1MOut: $0.0437/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
Opus 5.5
Anthropic
New
Opus 5.5 by Anthropic (2026-09-21): the next Opus, for long-running agentic coding and knowledge work. 1M context, 128K output, adaptive thinking always on.
1M
128K outIn: $0.0070/1MOut: $0.0350/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
Ossington 3
Augure
Large-context generalist with 128K window. Excellent for complex generation and analysis. Augure sovereign Canadian AI.
262K
33K outIn: $0.0013/1MOut: $0.0038/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamchattools
Try on Zubnet โ†’
Ossington 4.1
Augure
Premium multimodal model with vision and tool use, strong bilingual EN/FR. Augure sovereign Canadian AI.
262K
33K outIn: $0.0032/1MOut: $0.0076/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamchattoolsvisionimage
Try on Zubnet โ†’
Ossington 5
Augure
New
Newest reasoning model: the fastest of the Augure menu, strong at code, 256K context. Augure sovereign Canadian AI.
262K
33K outIn: $0.0019/1MOut: $0.0038/1MSpeed: โ—โ—โ—โ—โ—
Streamchatreasoning
Try on Zubnet โ†’
Qwen Flash
Alibaba Cloud
Ultra-fast model with 1M context. Best latency and cost efficiency for simple tasks.
1M
8K outIn: $0.000039/1MOut: $0.000378/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
Qwen Plus
Alibaba Cloud
Balanced flagship model with 1M context. Best performance/cost ratio for most tasks.
1M
8K outIn: $0.000700/1MOut: $0.0021/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
Qwen3 Coder Flash
Alibaba Cloud
Cost-effective coding model optimized for speed. Fast code generation and completion at lower cost.
131K
8K outIn: $0.000252/1MOut: $0.0010/1MSpeed: โ—โ—โ—โ—โ—
ToolsStreamtools
Try on Zubnet โ†’
Qwen3 VL Plus
Alibaba Cloud
Advanced vision-language model with 262K context. Excels at visual coding, spatial perception, and multimodal reasoning.
262K
8K outIn: $0.000250/1MOut: $0.0025/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
Qwen3.5 Flash
Alibaba Cloud
Flagship 1T+ parameter model. State-of-the-art reasoning, coding, and agent capabilities.
262K
33K outIn: $0.000051/1MOut: $0.000502/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
Qwen3.5 Omni Flash
Alibaba Cloud
Alibaba Qwen3.5 Omni Flash, fast multimodal (text & image input, text output). Singapore deployment; native voice/audio coming.
66K
8K outIn: $0.000700/1MOut: $0.0039/1MSpeed: โ—โ—โ—โ—โ—‹
Streamimage
Try on Zubnet โ†’
Qwen3.5 Omni Plus
Alibaba Cloud
Alibaba Qwen3.5 Omni, multimodal (text & image input, text output). Singapore deployment; native voice/audio coming.
66K
8K outIn: $0.0024/1MOut: $0.0145/1MSpeed: โ—โ—โ—โ—‹โ—‹
Streamimage
Try on Zubnet โ†’
Qwen3.5 Plus
Alibaba Cloud
397B MoE model with 17B active parameters, 1M token context, multimodal (text/image/video)
1M
8K outIn: $0.000201/1MOut: $0.0012/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimagevideo
Try on Zubnet โ†’
Qwen3.6 Flash
Alibaba Cloud
Flagship 1T+ parameter model. State-of-the-art reasoning, coding, and agent capabilities.
262K
33K outIn: $0.000289/1MOut: $0.0017/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
Qwen3.6 Plus
Alibaba Cloud
Qwen3.6 Plus, next-gen frontier LLM (Agentic Coding, vision/OCR, fine-grained localization). Replaces Qwen3.5 Plus.
1M
8K outIn: $0.000483/1MOut: $0.0029/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimagevideo
Try on Zubnet โ†’
Qwen3.7 Max
Alibaba Cloud
Alibaba's Agent Frontier flagship.
262K
33K outIn: $0.0029/1MOut: $0.0087/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
Qwen3.7 Plus
Alibaba Cloud
Multimodal frontier (vision/video); cheaper than Max.
1M
8K outIn: $0.000483/1MOut: $0.0019/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimagevideo
Try on Zubnet โ†’
Qwen3.8 Max
Alibaba Cloud
Alibaba's 2.4T-parameter multimodal MoE flagship. 1M context, accepts text and images.
1M
33K outIn: $0.0035/1MOut: $0.0105/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
Rosedale One
Augure
Premium model for complex, multi-step work. Augure sovereign Canadian AI.
1.048576M
33K outIn: $0.0032/1MOut: $0.0076/1MSpeed: โ—โ—โ—‹โ—‹โ—‹
ToolsStreamchattools
Try on Zubnet โ†’
Sarvam 105B
Sarvam AI
Sarvam's 24B multilingual LLM, fluent in all 22 official Indian languages + English. OpenAI-compatible API. Free tier with no per-token charges. Supports wiki grounding, reasoning effort control, and tool calls.
In: $0.000081/1MOut: $0.000325/1M
Toolstoolsreasoning
Try on Zubnet โ†’
Solar Mini
Upstage
Upstage lightweight model for fast, cost-effective inference.
33K
4K outIn: $0.000262/1MOut: $0.0010/1MSpeed: โ—โ—โ—โ—โ—
Streamchat
Try on Zubnet โ†’
Solar Pro 2
Upstage
Upstage previous-generation pro model with 64K context.
66K
8K outIn: $0.000262/1MOut: $0.0010/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamchattools
Try on Zubnet โ†’
Solar Pro 2 Nightly
Upstage
Upstage nightly build of Solar Pro 2 with latest improvements.
66K
8K outIn: $0.000262/1MOut: $0.0010/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamchattools
Try on Zubnet โ†’
Solar Pro 3
Upstage
Upstage flagship 100B MoE model with reasoning capabilities and strong multilingual performance.
131K
8K outIn: $0.000262/1MOut: $0.0010/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamchatreasoningtools
Try on Zubnet โ†’
Sonar
Perplexity
Perplexity's lightweight search-augmented model, real-time web search with citations, fast responses, 128K context.
128K
16K outIn: $0.0018/1MOut: $0.0018/1MSpeed: โ—โ—โ—โ—โ—
Streamsearch
Try on Zubnet โ†’
Sonar Deep Research
Perplexity
Perplexity's most thorough research model, multi-step deep web investigation with comprehensive citations and reasoning.
128K
33K outIn: $0.0035/1MOut: $0.0140/1MSpeed: โ—โ—‹โ—‹โ—‹โ—‹
Streamsearchreasoning
Try on Zubnet โ†’
Sonar Pro
Perplexity
Perplexity's advanced search model, deeper web research, multi-step reasoning with citations, 200K context.
200K
16K outIn: $0.0053/1MOut: $0.0262/1MSpeed: โ—โ—โ—โ—‹โ—‹
ToolsStreamtoolssearch
Try on Zubnet โ†’
Sonar Reasoning Pro
Perplexity
Perplexity's reasoning model, extended thinking with real-time web search, ideal for complex research and analysis.
128K
16K outIn: $0.0035/1MOut: $0.0140/1MSpeed: โ—โ—โ—‹โ—‹โ—‹
ToolsStreamtoolssearchreasoning
Try on Zubnet โ†’
Sonnet 4.5
Anthropic
Claude's best model for complex agents and coding
200K
64K outIn: $0.0053/1MOut: $0.0262/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
Sonnet 4.6
Anthropic
Best Claude Sonnet for complex agents and coding (1M context)
1M
64K outIn: $0.0053/1MOut: $0.0262/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
Sonnet 5
Anthropic
Anthropic's most capable Sonnet โ€” closes the gap with Opus for complex agents and coding (1M context)
1M
64K outIn: $0.0035/1MOut: $0.0175/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsreasoningimage
Try on Zubnet โ†’
Step 3.5 Flash
StepFun
StepFun's latest reasoning model, 196B MoE (11B active), 256K context, open-source Apache 2.0. Fast agentic intelligence with strong code and math.
256K
16K outIn: $0.000175/1MOut: $0.000525/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
Step 3.7 Flash
StepFun
StepFun's multimodal flagship reasoning model.
256K
16K outIn: $0.000350/1MOut: $0.0020/1MSpeed: โ—โ—โ—โ—โ—‹
ToolsStreamtoolsreasoning
Try on Zubnet โ†’
Syn Pro
Upstage
Upstage advanced synthetic reasoning model with 64K context.
66K
8K outIn: $0.000262/1MOut: $0.0010/1MSpeed: โ—โ—โ—โ—โ—‹
Streamchat
Try on Zubnet โ†’
Tofino 3
Augure
New
Light tier: fast, thinking off, 256K context, served in Canada with no failover outside Canada. Replaces Tofino 2.5, retired by Augure on 2026-09-19.
262K
33K outIn: $0.000756/1MOut: $0.0025/1MSpeed: โ—โ—โ—โ—โ—
Streamchat
Try on Zubnet โ†’
ESC
↑↓ Navigate ↵ Open wiki: news: model: Filter