The Stav model catalogue

Models

DeepSeek

Released 17 Oct 2026

DeepSeek's vision-capable sparse MoE model for text and image input, with a 1M-token context window and MIT-licensed open weights.

€0.45 / 1.35per 1M in / out

L11 provider

Anthropic

Claude Haiku 5.5

Released 7 Oct 2026

Anthropic's low-latency multimodal model featuring a 1,000,000-token context window and configurable reasoning effort.

€0.09 / 0.45per 1M in / out

L01 provider

Mistral AI

Mistral Large 4

Released 6 Oct 2026

Mistral AI's 1.05T MoE frontier multimodal model with 52B active parameters, 512K context, and configurable reasoning traces.

€0.67 / 2.06per 1M in / out

L11 provider

Page 1 of 2‹ PrevNext ›

Google

Gemini 4 Argon

Released 30 Sept 2026

Google's closed-weight Gemini 4 Argon is a multimodal reasoning model with configurable thinking levels and a 1,048,576-token context window.

€1.76 / 8.80per 1M in / out

L01 provider

OpenAI

gpt-6.1-sol

Released 29 Sept 2026

OpenAI's closed-weight reasoning model with a 1,050,000-token context window, mandatory configurable reasoning effort and native hosted tools, sitting between GPT-6 Sol and the GPT-6 Astra flagship.

€1.80 / 8.98per 1M in / out

L01 provider

Anthropic

Claude Sonnet 5.5

Released 28 Sept 2026

Anthropic's closed-weight Sonnet-class reasoning model with a 1M-token context window, always-on adaptive thinking, and configurable effort levels.

€1.80 / 8.98per 1M in / out

L01 provider

Anthropic

Claude Opus 5.5

Released 22 Sept 2026

Anthropic's flagship hybrid reasoning model, with always-on adaptive thinking and a 1M-token context window for deep analytical and coding work.

€3.59 / 17.95per 1M in / out

L01 provider

OpenAI

gpt-6-luna

Released 22 Sept 2026

OpenAI's fast, cost-efficient GPT-6 model for high-volume chat, classification, and lightweight agentic tasks, with dialable reasoning effort and a 1.05M-token context window.

€0.09 / 0.45per 1M in / out

L01 provider

OpenAI

gpt-6-sol

Released 22 Sept 2026

OpenAI's mid-tier reasoning model with configurable effort levels, a 1.05M-token context window, and text+image input for coding and agentic workflows.

€1.80 / 8.98per 1M in / out

L01 provider

Xiaomi MiMo

MiMo-V2.6-Flash

Released 22 Sept 2026

Xiaomi's MIT-licensed efficiency-tier omnimodal reasoning model: a 309B/15B sparse MoE with a 1M-token context, native image, video and audio understanding, and a caller-controlled thinking switch.

€0.13 / 0.25per 1M in / out

L01 provider

xAI

grok-4.7

Released 21 Sept 2026

xAI's reasoning-first flagship with a 500K-token context window, built for deep coding, agentic, and long-document knowledge work.

€1.80 / 5.39per 1M in / out

L01 provider

Xiaomi MiMo

MiMo-V2.6-Pro

Released 21 Sept 2026

Xiaomi's open-weight 1.02T-parameter omnimodal reasoning model with a 1M-token context, accepting text, image, video and audio input and returning text.

€0.39 / 0.78per 1M in / out

L01 provider

DeepSeek

DeepSeek V4.1 Flash

Released 10 Sept 2026

DeepSeek's unified multimodal MoE model combining native text+image input, a 1M-token context window, and a controllable reasoning-effort slider in one Flash-tier endpoint.

€0.18 / 0.54per 1M in / out

L13 providers

OpenAI

gpt-6-astra

Released 3 Sept 2026

OpenAI's flagship closed-weight reasoning model with a 1.05M-token context window and mandatory reasoning on every request.

€8.98 / 44.88per 1M in / out

L01 provider

Google

Gemini 3.8 Flash

Released 2 Sept 2026

Google's reasoning-tuned Flash-tier model for fast coding, agentic, and multi-step reasoning tasks with a million-token multimodal context window.

€0.66 / 3.30per 1M in / out

L01 provider

Anthropic

Claude Fable 5.1

Released 1 Sept 2026

Anthropic's frontier reasoning model with a 1M-token context window, always-on adaptive reasoning, and native text-and-image input.

€8.98 / 44.88per 1M in / out

L01 provider

Z.ai

GLM-5.3 Flash

Released 26 Aug 2026

Z.ai's natively multimodal Mixture-of-Experts model with a 1M+ token context window and mandatory reasoning.

€0.13 / 0.45per 1M in / out

L15 providers

Z.ai

GLM-5.3

Released 14 Aug 2026

Z.ai's flagship 743B-parameter MoE model, open-weighted and post-trained for complex coding and long-horizon agentic tasks, with a 1M-token context window.

€0.81 / 3.59per 1M in / out

L36 providers

DeepSeek

DeepSeek V4 Pro (0813)

Released 13 Aug 2026

DeepSeek's open-weight, MIT-licensed V4 Pro reasoning model — a large mixture-of-experts system with a million-token context window and selectable reasoning effort for long, multi-step tasks.

€1.17 / 2.33per 1M in / out

L01 provider

xAI

grok-4.6

Released 12 Aug 2026

xAI's closed-weight, reasoning-tuned model for long-running agentic coding and research, with a 500K-token context window and configurable reasoning depth.

€1.80 / 5.39per 1M in / out

L01 provider

Qwen

Qwen3.8 2.4T A95B

Released 12 Aug 2026

Alibaba/Qwen's largest open-weight release: a 2.4T-parameter mixture-of-experts model with mandatory reasoning and a 262K-token context window.

€1.80 / 5.39per 1M in / out

L13 providers

Qwen

Qwen 3.8 Max

Released 12 Aug 2026

Alibaba's open-weight, always-reasoning Qwen3.8 Max-class mixture-of-experts model with a 262K-token native context window.

€1.80 / 5.39per 1M in / out

L01 provider

NVIDIA

Nemotron Lightning 3.5 30B A3B

Released 11 Aug 2026

An open-weights, low-latency NVIDIA reasoning and agentic model with a toggleable chain-of-thought mode and a hybrid Mamba-2/MoE/Attention architecture.

€0.04 / 0.18per 1M in / out

L01 provider

DeepSeek

DeepSeek V4 Flash (0731)

Released 31 Jul 2026

A million-token-context, mixture-of-experts reasoning model from DeepSeek with adjustable thinking effort, served on the Stav Platform via Fireworks AI.

€0.05 / 0.16per 1M in / out

L33 providers

Anthropic

Claude Opus 5

Released 24 Jul 2026

Anthropic's frontier reasoning model with a 1M-token context, adjustable reasoning effort, and multimodal input for code, research, and long-document work.

€4.49 / 22.44per 1M in / out

L01 provider

Google

Gemini 3.5 Flash Lite

Released 21 Jul 2026

Google's fast, cost-focused Gemini 3.5 model with a million-token context window, mandatory configurable reasoning, and multimodal text, image, audio, and video input.

€0.26 / 2.20per 1M in / out

L01 provider

Google

Gemini 3.6 Flash

Released 21 Jul 2026

Google's reasoning-tuned Flash model: multimodal input, a million-token-plus context window, and always-on chain-of-thought for agentic, long-horizon tasks.

€0.66 / 3.30per 1M in / out

L01 provider

Moonshot AI

Kimi K3

Released 16 Jul 2026

Moonshot AI's open-weights, always-reasoning 2.8T MoE flagship with a 1M-token context, native tool use, and multimodal input.

€2.56 / 12.79per 1M in / out

L25 providers

Thinking Machines Lab

Inkling

Released 15 Jul 2026

Thinking Machines Lab's first open-weight model: a 975B/41B-active MoE transformer with multimodal input, a 1M-token context window, and configurable reasoning effort.

€0.90 / 3.64per 1M in / out

L01 provider

OpenAI

gpt-5.6-luna

Released 9 Jul 2026

OpenAI's fast, cost-efficient GPT-5.6 tier for high-volume chat, classification, and lightweight agentic tasks with a 1.05M-token context window.

€0.18 / 1.08per 1M in / out

L01 provider

OpenAI

gpt-5.6-sol

Released 9 Jul 2026

OpenAI's flagship GPT-5.6 reasoning model for agentic coding, cybersecurity, and long-horizon analysis, with a 1.05M-token context window.

€3.59 / 17.95per 1M in / out

L01 provider

OpenAI

gpt-5.6-terra

Released 9 Jul 2026

OpenAI's balanced GPT-5.6 model, offering a 1.05M-token context window and configurable reasoning effort for mid-complexity coding and document tasks.

€1.80 / 10.77per 1M in / out

L01 provider

xAI

grok-4.5

Released 8 Jul 2026

xAI's proprietary reasoning model with a 500,000-token context window, mandatory configurable reasoning effort, and text and image input.

€1.80 / 5.39per 1M in / out

L01 provider

Anthropic

Claude Sonnet 5

Released 30 Jun 2026

Anthropic's adaptive-reasoning Sonnet-class model with a 1M-token context window, strong Design Arena coding and UI results, and always-on reasoning selectable from low to max effort.

€1.80 / 8.98per 1M in / out

L01 provider

Z.ai

glm-5-2

Released 13 Jun 2026

Z.ai's flagship open-weights reasoning and agentic-coding model, with a 1M-token context window and MIT-licensed weights.

€0.67 / 2.15per 1M in / out

L35 providers

Moonshot AI

MoonshotAI: Kimi K2.7 Code

Released 12 Jun 2026

Moonshot AI's 1T-parameter open-weights MoE coding and agentic model with mandatory extended thinking and a 262,144-token context window.

€1.15 / 4.05per 1M in / out

L11 provider

Anthropic

Claude Fable 5

Released 9 Jun 2026

Anthropic's frontier Mythos-class model with always-on adaptive reasoning and a one-million-token context window, built for deep code and research analysis rather than real-time chat.

€8.98 / 44.88per 1M in / out

L01 provider

NVIDIA

Nemotron 3 Ultra (NVFP4)

Released 4 Jun 2026

NVIDIA's flagship 550B/55B-active Nemotron 3 Ultra reasoning model, open-weighted under OpenMDW-1.1 with a 256K-token context window.

€0.54 / 2.15per 1M in / out

L01 provider

MiniMax

MiniMax: MiniMax M3

Released 1 Jun 2026

An open-weight 428B-parameter (~23B active) MoE model from MiniMax with a 1M-token context window, multimodal input, and a configurable thinking mode, suited to long-context, agentic, and coding-assist workloads.

€0.27 / 1.08per 1M in / out

L01 provider

xAI

grok-build-0.1

Released 29 May 2026

xAI's agentic coding model: a 256K-context, tool-calling assistant with mandatory step-by-step reasoning, built for multi-step debugging and MCP orchestration.

€0.90 / 1.80per 1M in / out

L01 provider

Anthropic

Claude Opus 4.8

Released 28 May 2026

Anthropic's frontier reasoning model with a 1M-token context window, built for complex agentic workflows, graduate-level analysis, and long-context tasks.

€4.49 / 22.44per 1M in / out

L01 provider

Mistral AI

mistral-medium-3.5-128b

Released 22 May 2026

Dense 128B open-weight multimodal model from Mistral AI with a 256K context and optional reasoning.

€1.50 / 7.50per 1M in / out

L31 provider

Google

Gemini 3.1 Flash Lite

Released 7 May 2026

Google's fast, cost-optimised multimodal Gemini model with a million-token context window, built for high-throughput text, image, audio, video, and file input.

€0.22 / 1.32per 1M in / out

L01 provider

xAI

grok-4.3

Released 30 Apr 2026

xAI's reasoning model with a 1M-token context window and configurable reasoning effort, on by default.

€1.12 / 2.24per 1M in / out

L01 provider

OpenAI

gpt-5.5

Released 23 Apr 2026

OpenAI's long-context instruct model, built for complex multi-step reasoning, long-document analysis, and tool orchestration across a 1M+ token context window.

€4.49 / 26.93per 1M in / out

L01 provider

Anthropic

Claude Opus 4.7

Released 16 Apr 2026

Anthropic's frontier-tier closed-weight instruct model with a one-million-token context window, built for complex, long-horizon reasoning and coding rather than real-time interaction.

€4.49 / 22.44per 1M in / out

L01 provider

Z.ai

GLM-5.1

Released 7 Apr 2026

Z.ai's 754B-parameter open-weight MoE model with integrated reasoning, built for long-horizon agentic engineering and state-of-the-art coding benchmark performance.

€1.26 / 3.95per 1M in / out

L22 providers

Google

Gemma 4 26B A4B It

Released 2 Apr 2026

Google DeepMind's open-weight, multimodal Mixture-of-Experts model with a 262K-token context window, tuned for interactive-latency, moderate-complexity tasks.

Free / Freeper 1M in / out

L34 providers