From ff3879a39cd673e6a39a17d0a5b88b766e3737b4 Mon Sep 17 00:00:00 2001 From: Guilherme Soubihe Date: Thu, 22 Jan 2026 17:40:47 -0300 Subject: [PATCH 1/3] Add Latitude.sh as inference provider Latitude.sh is a bare-metal cloud provider with 10,000+ physical servers and 1,000+ GPUs across global data centers. The AI inference platform provides OpenAI-compatible API with competitive pricing. Features: - Tool calling and structured output (JSON mode) - Vision/multimodal model support - Streaming responses - Dedicated GPUs with consistent performance Co-Authored-By: Claude Opus 4.5 --- .../templates/providers/latitude.handlebars | 74 +++++++++++++++++++ 1 file changed, 74 insertions(+) create mode 100644 scripts/inference-providers/templates/providers/latitude.handlebars diff --git a/scripts/inference-providers/templates/providers/latitude.handlebars b/scripts/inference-providers/templates/providers/latitude.handlebars new file mode 100644 index 0000000000..087f471cc9 --- /dev/null +++ b/scripts/inference-providers/templates/providers/latitude.handlebars @@ -0,0 +1,74 @@ +# Latitude: Bare-Metal Cloud & AI Inference + +> [!TIP] +> All supported Latitude models can be found [here](https://huggingface.co/models?inference_provider=latitude-sh&sort=trending) + +{{{logoSection}}} + +{{{followUsSection}}} + +## Fast, reliable AI inference powered by dedicated bare-metal GPU infrastructure. + +[Latitude.sh](https://latitude.sh) is a leading bare-metal cloud provider operating **10,000+ physical servers** and **1,000+ GPUs** across a global network of data centers. Our infrastructure powers mission-critical workloads for enterprises, AI startups, and developers worldwide. + +--- + +## Global Presence + +- πŸ‡ΊπŸ‡Έ **United States**: Dallas, Chicago, New York, Miami, Los Angeles, Seattle +- πŸ‡§πŸ‡· **Brazil**: SΓ£o Paulo +- πŸ‡¬πŸ‡§ **United Kingdom**: London +- πŸ‡©πŸ‡ͺ **Germany**: Frankfurt +- πŸ‡³πŸ‡± **Netherlands**: Amsterdam +- πŸ‡―πŸ‡΅ **Japan**: Tokyo +- πŸ‡ΈπŸ‡¬ **Singapore** +- πŸ‡¦πŸ‡Ί **Australia**: Sydney + +--- + +## Infrastructure + +- **1,000+ GPUs** including NVIDIA H100, A100, L40S, and RTX PRO 6000 Blackwell +- **10,000+ physical servers** (CPU and GPU) +- Direct peering with **Megaport** for low-latency enterprise connectivity +- Tier-3 data centers with **99.99% SLA** + +--- + +## AI Inference Platform + +Our inference platform offers: + +- **OpenAI-compatible API**: Drop-in replacement for existing integrations +- **Dedicated GPUs**: Consistent performance without noisy neighbor issues +- **Competitive pricing**: Direct GPU ownership means better margins passed to you +- **Full feature support**: Tool calling, structured output (JSON mode), vision/multimodal inputs, and streaming + +--- + +## Pricing + +For the latest rates, visit our [pricing page](https://api.lsh.ai/pricing). + +| Model | Input ($/1M tokens) | Output ($/1M tokens) | +|-------|---------------------|----------------------| +| Llama 3.1 8B | $0.16 | $0.16 | +| Qwen 2.5 7B | $0.27 | $0.27 | +| Qwen 2.5 VL 7B (Vision) | $0.36 | $0.36 | +| Gemma 2 27B | $0.45 | $0.45 | +| Qwen3 32B | $0.45 | $0.90 | +| Qwen2.5 Coder 32B | $0.72 | $0.72 | +| DeepSeek R1 Distill 14B | $1.44 | $1.44 | + +--- + +## Resources + +- **Website**: [latitude.sh](https://latitude.sh) +- **API Documentation**: [api.lsh.ai/docs](https://api.lsh.ai/docs) +- **X (Twitter)**: [@lataboratories](https://x.com/lataboratories) +- **LinkedIn**: [Latitude.sh](https://www.linkedin.com/company/latitudeai/) +- **Discord**: [Join our community](https://discord.gg/latitude) +- **GitHub**: [@latitude-sh](https://github.com/latitude-sh) + +{{{tasksSection}}} From b62e5e48aecb1492d317c89746bba6a433afb619 Mon Sep 17 00:00:00 2001 From: Guilherme Soubihe Date: Thu, 22 Jan 2026 17:42:25 -0300 Subject: [PATCH 2/3] Update: Latitude is a Megaport company Co-Authored-By: Claude Opus 4.5 --- .../templates/providers/latitude.handlebars | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/scripts/inference-providers/templates/providers/latitude.handlebars b/scripts/inference-providers/templates/providers/latitude.handlebars index 087f471cc9..0659b24bc8 100644 --- a/scripts/inference-providers/templates/providers/latitude.handlebars +++ b/scripts/inference-providers/templates/providers/latitude.handlebars @@ -9,7 +9,7 @@ ## Fast, reliable AI inference powered by dedicated bare-metal GPU infrastructure. -[Latitude.sh](https://latitude.sh) is a leading bare-metal cloud provider operating **10,000+ physical servers** and **1,000+ GPUs** across a global network of data centers. Our infrastructure powers mission-critical workloads for enterprises, AI startups, and developers worldwide. +[Latitude.sh](https://latitude.sh), a **Megaport company**, is a leading bare-metal cloud provider operating **10,000+ physical servers** and **1,000+ GPUs** across a global network of data centers. Our infrastructure powers mission-critical workloads for enterprises, AI startups, and developers worldwide. --- @@ -30,7 +30,7 @@ - **1,000+ GPUs** including NVIDIA H100, A100, L40S, and RTX PRO 6000 Blackwell - **10,000+ physical servers** (CPU and GPU) -- Direct peering with **Megaport** for low-latency enterprise connectivity +- Part of **Megaport's global network** for low-latency enterprise connectivity - Tier-3 data centers with **99.99% SLA** --- From 777829e12fda5c3babb9a0fcd9ee32eafaebd104 Mon Sep 17 00:00:00 2001 From: Guilherme Soubihe Date: Thu, 22 Jan 2026 17:43:45 -0300 Subject: [PATCH 3/3] Update: Latitude acquired by Megaport (ASX: MP1) Reflect the November 2025 acquisition and combined platform capabilities. Co-Authored-By: Claude Opus 4.5 --- .../templates/providers/latitude.handlebars | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/scripts/inference-providers/templates/providers/latitude.handlebars b/scripts/inference-providers/templates/providers/latitude.handlebars index 0659b24bc8..9e957b9ece 100644 --- a/scripts/inference-providers/templates/providers/latitude.handlebars +++ b/scripts/inference-providers/templates/providers/latitude.handlebars @@ -9,7 +9,7 @@ ## Fast, reliable AI inference powered by dedicated bare-metal GPU infrastructure. -[Latitude.sh](https://latitude.sh), a **Megaport company**, is a leading bare-metal cloud provider operating **10,000+ physical servers** and **1,000+ GPUs** across a global network of data centers. Our infrastructure powers mission-critical workloads for enterprises, AI startups, and developers worldwide. +[Latitude.sh](https://latitude.sh), recently **acquired by [Megaport](https://megaport.com)** (ASX: MP1), is a leading bare-metal cloud provider operating **10,000+ physical servers** and **1,000+ GPUs** across a global network of data centers. Together with Megaport's Network-as-a-Service platform spanning 1,000+ data centers in 26 countries, we power mission-critical workloads for enterprises, AI startups, and developers worldwide. --- @@ -30,7 +30,7 @@ - **1,000+ GPUs** including NVIDIA H100, A100, L40S, and RTX PRO 6000 Blackwell - **10,000+ physical servers** (CPU and GPU) -- Part of **Megaport's global network** for low-latency enterprise connectivity +- Combined with **Megaport's global NaaS platform** spanning 1,000+ data centers in 26 countries - Tier-3 data centers with **99.99% SLA** ---