Every model, one endpoint
Browse 36+ open and frontier models on the Parasail inference cloud. Filter by category, compare specs, and call any model with a single OpenAI-compatible API.
Featured models
View all →GLM-5.2
Z.AI’s next-gen model for agentic engineering with sustained execution on long-horizon tasks.
GLM-5.1
Strong reasoning and coding model with a 1M context window at competitive pricing.
MiniMax M3
Highly efficient MoE model delivering strong reasoning at exceptionally low cost.
DeepSeek V4 Pro
Open model for reasoning and coding workloads.
Kimi K2.7 Code
Specialized coding model optimized for software engineering and agentic coding workflows.
Qwen3-VL 235B-A22B
Large vision-language model with strong document understanding and visual reasoning.
Recently added
View all →Qwen3.6 35B-A3B
Efficient MoE model with 3B active parameters — great price-to-performance.
Nemotron 3 Ultra 550B
NVIDIA’s flagship open model optimized for enterprise reasoning workloads.
GLM-5.2
Z.AI’s next-gen model for agentic engineering with sustained execution on long-horizon tasks.
Kimi K2.7 Code
Specialized coding model optimized for software engineering and agentic coding workflows.
Explore the library
Filter by category to find the right model for your use case.
GLM-5.2
Z.AI’s next-gen model for agentic engineering with sustained execution on long-horizon tasks.
GLM-5.1
Strong reasoning and coding model with a 1M context window at competitive pricing.
GLM-5
Z.AI’s open model for reasoning and agentic workflows.
MiniMax M3
Highly efficient MoE model delivering strong reasoning at exceptionally low cost.
MiniMax M2.5
Cost-effective large model with strong general capabilities and long context.
DeepSeek V4 Pro
Open model for reasoning and coding workloads.
DeepSeek V4 Flash
Ultra-fast, ultra-cheap reasoning model for high-throughput workloads.
Kimi K2.6
Strong reasoning model with excellent long-context understanding.
Qwen3.5 397B-A17B
Large-scale MoE model with frontier reasoning and multilingual capabilities.
Qwen3 235B-A22B
Strong open MoE model with excellent coding and reasoning at low cost.
Trinity Large (Thinking)
Reasoning-focused model with extended thinking for complex problem-solving.
Nemotron 3 Ultra 550B
NVIDIA’s flagship open model optimized for enterprise reasoning workloads.
Llama 4 Maverick
Meta’s multimodal Llama model with native image understanding.
Llama 3.3 70B
Meta’s reliable workhorse — great quality-to-cost ratio for production.
Kimi K2.7 Code
Specialized coding model optimized for software engineering and agentic coding workflows.
Qwen3-Coder-Next
Code-specialized model with strong fill-in-the-middle and agentic coding support.
Qwen3-VL 235B-A22B
Large vision-language model with strong document understanding and visual reasoning.
Qwen3-VL 8B
Compact vision-language model — fast and affordable for image understanding at scale.
Qwen2.5-VL 72B
Proven vision-language model for OCR, document QA, and visual grounding.
Qwen3.6 35B-A3B
Efficient MoE model with 3B active parameters — great price-to-performance.
Qwen3.5 35B-A3B
Compact MoE model with strong reasoning at minimal cost per token.
Qwen3-Next 80B
Balanced model with strong general capabilities at a competitive price point.
Mistral Small 3.2 24B
Mistral’s efficient model — excellent for high-volume, latency-sensitive workloads.
Gemma 4 26B-A4B
Google’s efficient MoE Gemma model with 4B active parameters.
Gemma 4 31B
Google’s latest open model with strong general performance and broad language support.
Gemma 3 27B
Proven compact model with multimodal capabilities and strong multilingual support.
gpt-oss-120b
OpenAI’s open-weight model with strong reasoning at a low cost.
gpt-oss-120b (Fast)
Optimized fast tier of gpt-oss-120b with lower latency for real-time workloads.
gpt-oss-20b
Compact open-weight model from OpenAI — ultra-cheap, fits on a single GPU.
MiMo v2.5
Xiaomi’s efficient reasoning model with strong coding capabilities.
Skyfall 31B v4.2
Fine-tuned model optimized for creative writing and conversational tasks.
Skyfall 36B v2
Fine-tuned model with enhanced instruction-following and creative output.
Cydonia 24B v4.1
Fine-tuned model with enhanced roleplay and character consistency.
BGE-M3
Multilingual, multi-functional embedding model supporting 100+ languages.
Resemble TTS (English)
High-quality text-to-speech model with natural-sounding English voices.
UI-TARS 1.5 7B
GUI agent model that can operate computer interfaces autonomously.