16 models, one API key
Every model is served over an OpenAI-compatible endpoint and priced per token. Filter by capability, copy the model id, and start calling Speka in minutes.
DeepSeek V4 Pro
State-of-the-art open reasoning model with transparent chain-of-thought. Excels at math, logic and multi-step problem solving.
deepseek-ai/deepseek-v4-pro
Nemotron Super 49B
NVIDIA's reasoning-tuned Nemotron model — strong math, logic and agentic tool-use at an efficient size.
nvidia/llama-3.3-nemotron-super-49b-v1.5Nemotron Nano 9B
Compact reasoning model that punches above its weight on math and coding benchmarks. Fast and cheap.
nvidia/nvidia-nemotron-nano-9b-v2
Llama 3.3 70B Instruct
Meta's flagship 70B instruct model — 405B-class quality at a fraction of the cost. Great default for production chat.
meta/llama-3.3-70b-instructLlama 4 Maverick
Meta's latest flagship mixture-of-experts model. Frontier quality for demanding reasoning and generation.
meta/llama-4-maverick-17b-128e-instructLlama 3.1 8B Instruct
Fast, cheap and capable. Ideal for high-volume classification, routing and lightweight chat.
meta/llama-3.1-8b-instruct
Mistral Large 3
Mistral's top-tier model with excellent reasoning, function calling and 80+ language support.
mistralai/mistral-large-3-675b-instruct-2512Mistral Medium 3.5
Balanced mid-size Mistral model with strong multilingual and coding ability at efficient cost.
mistralai/mistral-medium-3.5-128b
Qwen3 Coder 480B
Best-in-class open code model. Excellent at generation, completion, refactoring and bug fixing.
qwen/qwen3-coder-480b-a35b-instructLlama 3.2 90B Vision
Large multimodal model for image understanding, document Q&A, charts and visual reasoning.
meta/llama-3.2-90b-vision-instructLlama 3.2 11B Vision
Lightweight vision-language model for fast image captioning, OCR and visual Q&A.
meta/llama-3.2-11b-vision-instruct
Phi-4 Multimodal
Microsoft's compact multimodal model handling text, image and audio understanding.
microsoft/phi-4-multimodal-instructNV-EmbedQA E5 v5
Robust retrieval embeddings tuned for question answering and RAG pipelines.
nvidia/nv-embedqa-e5-v5NV-Embed v1
High-accuracy general-purpose text embeddings for semantic search and clustering.
nvidia/nv-embed-v1
FLUX.1 [dev]
High-fidelity text-to-image generation with excellent prompt adherence and typography.
black-forest-labs/flux.1-dev
Stable Diffusion 3.5 Large
Stability's flagship image model for photorealistic and artistic generation.
stabilityai/stable-diffusion-3.5-large