We just added 16 new AI models — Claude Sonnet 5.5, GPT-6.1 Sol, Gemini 4 Argon & more

We have just added sixteen new LLM chat models to YepAPI. Anthropic's Claude Sonnet 5.5 and Haiku 5.5, OpenAI's GPT-6.1 Sol, and Google's first Gemini 4 model, Gemini 4 Argon, lead the batch, alongside Mistral Large 4, Moonshot's Kimi K3, two new Meta Muse models and a new provider, Inception, with its diffusion-based Mercury 3. Every model is callable from the same `/v1/ai/chat` (and OpenAI-compatible `/v1/ai/chat/completions`) endpoint under the same key, with pay-per-token pricing.

What's new

Anthropic Claude Sonnet 5.5 ($2.95 / $14.77) and Claude Haiku 5.5 ($0.15 / $0.74), both with a 1M-token context
OpenAI GPT-6.1 Sol and GPT-6.1 Sol Pro, the upgrade to GPT-6 Sol
Google Gemini 4 Argon, the first Gemini 4 model, with text, image, audio, video and file input
Mistral Large 4, Moonshot Kimi K3, Meta Muse Wonder and Muse Spark Open
MiniMax M3.1 Flash, Qwen3.7 Plus and Flash, StepFun Step 5 Preview, NVIDIA Nemotron 3.5 Super VL and DeepSeek R1 0528
New provider: Inception, with Mercury 3, a diffusion language model built for low latency
Aliases: `claude-sonnet` → Sonnet 5.5, `claude-haiku` → Haiku 5.5, `deepseek-r1` → R1 0528; new `gpt-6.1`, `gemini-4`, `kimi-k3`, `mistral-large` and `mercury`

Endpoints in this release

AIClaude Sonnet 5.5$2.95 / $14.77 per 1M tokens AIClaude Haiku 5.5$0.15 / $0.74 per 1M tokens AIGPT-6.1 Sol$2.95 / $14.77 per 1M tokens AIGPT-6.1 Sol Pro$2.95 / $14.77 per 1M tokens AIGemini 4 Argon$2.95 / $14.77 per 1M tokens AIMistral Large 4$2.01 / $6.17 per 1M tokens AIKimi K3$4.43 / $22.15 per 1M tokens AIMuse Wonder$5.91 / $29.54 per 1M tokens AIMuse Spark Open$1.85 / $6.28 per 1M tokens AIMiniMax M3.1 Flash$0.22 / $0.89 per 1M tokens AIQwen3.7 Plus$0.47 / $1.89 per 1M tokens AIQwen3.7 Flash$0.04 / $0.19 per 1M tokens AIStep 5 Preview$1.48 / $3.99 per 1M tokens AIMercury 3$0.15 / $0.52 per 1M tokens AINemotron 3.5 Super VL$0.22 / $0.74 per 1M tokens AIDeepSeek R1 0528$0.74 / $3.18 per 1M tokens

What's new

Claude Sonnet 5.5 is a direct upgrade to Sonnet 5 at the same price, and Claude Haiku 5.5 brings the Claude 5.5 generation to a price a fraction of Haiku 4.5. Gemini 4 Argon is Google's new flagship reasoning model and takes every input type on the platform. Kimi K3 is a 2.8T-parameter open-weight reasoning model, and Mercury 3 generates text by diffusion rather than one token at a time, which makes it very fast for agent loops.

Aliases updated

`claude-sonnet` now resolves to Claude Sonnet 5.5 (same price as Sonnet 5), `claude-haiku` to Claude Haiku 5.5 (cheaper than Haiku 4.5), and `deepseek-r1` to DeepSeek R1 0528, because the original R1 is no longer served. Pin the full model ID, for example `anthropic/claude-sonnet-5`, to stay on a specific release.

How to call them

Set the model field to the full ID or a short alias and everything else stays the same.

curl -X POST https://api.yepapi.com/v1/ai/chat \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "claude-sonnet-5.5", "messages": [{"role": "user", "content": "Summarize the trade-offs of event sourcing."}]}'