Geodd Infrastructure
Geodd is an AI inference company that serves models through one API and develops hardware-specific kernels to run those models more efficiently. It brings inference operations and kernel development together so real production workloads drive performance improvements.
At a Glance
- AI product and application developers
- AI teams deploying production inference
- Enterprise customers requiring custom scale, SLAs, or multi-region clusters
- Teams needing dedicated GPU or single-tenant inference capacity
- +2 more
AI Tools by Geodd Infrastructure
(1)
Geodd
AI Model Inference API
Discussions
No discussions yet
Be the first to start a discussion about Geodd Infrastructure
Latest News
7 Best OpenAI API Alternatives for Developers in 2026
ByteDance Seed, Seedream and Seedance Models Are Now Live on Geodd
Geodd's GDPR-Ready Approach to AI Inference
Opper AI and Geodd Partner to Add Production Ready Inference for AI Teams
Products & Services
Managed, usage-priced inference through a unified OpenAI-compatible API, including text, embedding, image-generation, and video-generation workflows.
Dedicated or isolated inference deployments for production workloads, with custom scale, enterprise support, and tailored infrastructure options.
Dedicated GPU infrastructure that can be prepared, created, connected to, and managed through the console and API.
Hardware-specific kernel-development system powered by specialist models: Mosaic for NVIDIA/CUDA, Druze for AMD/ROCm, and Strata for Tenstorrent/TT-Metalium.
Market Position
Geodd positions itself as an inference operator spanning physical data centers, GPU operations, model deployment, and AI-driven optimization. Its differentiation is the combination of serverless and dedicated inference with production-observed, hardware-specific kernel development, regional deployment, and direct engineering support; it is compared publicly with inference platforms such as Fireworks AI and VESSL AI.
Leadership
Founders
Malith Hatanachchige
Founder and CEO of Geodd; LinkedIn identifies him as based in Sri Lanka and F6S lists him as Geodd's founder. Search results also identify him as the former CEO and co-founder of Revox.
Executive Team
Malith Hatanachchige
Founder & CEO
Founder and CEO of Geodd; previously CEO and co-founder of Revox according to public profile search results.
Shani Kahathuduwa
Chief Product Officer
Crunchbase lists her as Geodd's Chief Product Officer and identifies a previous role as Vice President of Product at oDoc. LinkedIn describes her Geodd function as Product & Operations.
Founding Story
Geodd was established in 2023 to run production AI inference and make it more efficient, reliable, and predictable. Its initial vision connects operating inference systems with developing accelerator-specific kernels, using measured production bottlenecks to guide optimization work.
Business Model
Revenue Model
Usage-based token pricing for serverless inference, plus flexible pricing for dedicated inference, GPU instances, bare-metal infrastructure, and custom enterprise deployments. Geodd also offers volume discounts and tailored infrastructure solutions for large-scale deployments.
Pricing Tiers
Pricing is model-specific and usage-based; the pricing calculator shows a sample $290 monthly estimate for 1B input tokens and 500M output tokens.
Model listing price shown on Geodd's home page.
Model listing price shown on Geodd's home page.
Target Markets
- AI product and application developers
- AI teams deploying production inference
- Enterprise customers requiring custom scale, SLAs, or multi-region clusters
- Teams needing dedicated GPU or single-tenant inference capacity
- Organizations with EU/EEA/UK privacy and data-residency requirements
- Developers building multimodal text, image, and video applications
- AI agents and long-horizon tasks
- Coding, reasoning, research, tool-use, and information-extraction applications
- Production language-model APIs
- Image generation, editing, advertising, e-commerce, and marketing assets
- Video generation, social content, product videos, and automated creative pipelines
- High-throughput batch processing and latency-sensitive production APIs