sllm
Split a GPU node with other developers, unlimited tokens
Details
- External ID
- 47639779
- Source
- HN
- Company
- —
- Product
- sllm
- Website domain
- sllm.cloud
- Launched
- April 4, 2026
- Cohort
- —
- Upvotes
- 188
- Upvotes percentile
- 0.9550128534704371
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:26 p.m.
- Updated at
- Sept. 7, 2026, 9:26 p.m.
Description
Running DeepSeek V3 (685B) requires 8×H100 GPUs which is about $14k/month. Most developers only need 15-25 tok/s. sllm lets you join a cohort of developers sharing a dedicated node. You reserve a spot with your card, and nobody is charged until the cohort fills. Prices start at $5/mo for smaller models.The LLMs are completely private (we don't log any traffic).The API is OpenAI-compatible (we run vLLM), so you just swap the base URL. Currently offering a few models.
Enrichment
- Theme
- DeepSeek model deployment and inference
- Vertical
- Horizontal
- Function
- Model & infra
- Audience
- Developer
- AI stance
- AI-native
- Project type
- Commercial product
- Normalized one-liner
- shared gpu compute for developers
- Manually corrected
- False
Could you build this?
No Operating shared multi-tenant 8xH100 clusters with fair scheduling, custom GPU slicing, secure memory isolation, and high-throughput inference for 685B parameter models requires massive capital and specialized systems/GPU infra engineering.
What it would actually take: Building this requires a fleet orchestration layer on bare-metal GPU clusters (8xH100 NVLink nodes) using custom inference serving frameworks (vLLM, TensorRT-LLM) modified for fair-share token throttling and multi-tenant batch scheduling. It also needs a low-latency proxy layer with distributed billing/locking (Stripe integration, Redis locks) and secure execution isolation to prevent noisy-neighbor memory exhaustion, requiring expert GPU cluster and systems engineers.
Discussion
20 comments analyzed.
Competitors mentioned: OpenRouter, Chutes, Novita, hotaisle.xyz
Concerns raised: Product shutting down, Support email bouncing/unresponsive, Difficult to unsubscribe or close account, Earlier cohorts being canceled, Less flexible than competitors (can't switch models per-request)
Feature requests: 1-week or 1-day trial option before monthly commitment, Support for additional models beyond Qwen (MiMo-V2-Pro, Trinity), Ability to switch between models per-request, Stronger TEE guarantees if threat model requires it
Competitors
Other products that read as similar to this one — 224 launches clear the similarity bar, closest 8 shown.
Attention rank: #8 of 225 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 151 days after the earliest competitor.
- GOSH.AI DePools · ph · 2026-09-25 · 1 upvotes · similarity 0.52
- deepseekv4pro-deepseek-v4-pro-api-pricing · github · 2026-09-24 · 58 upvotes · similarity 0.50
- LumaDock - Blackwell GPU VPS · ph · 2026-09-14 · 2 upvotes · similarity 0.46
- deepseekv4flash-deepseek-v4-flash-api · github · 2026-09-24 · 54 upvotes · similarity 0.44
- Kohak AI · ph · 2026-09-26 · 2 upvotes · similarity 0.44
- DeepSeek Flash inverted the economics of agent products · hn · 2026-06-25 · 9 upvotes · similarity 0.44
- deepseek-v4.1-flash-next-dgx-spark-512k · github · 2026-09-13 · 6 upvotes · similarity 0.44
- cheapest-llm-api-de · github · 2026-09-28 · 26 upvotes · similarity 0.44
Other launches for this product
- No other launches for this product.
Same idea, different domain
Nobody's really built a model & infra tool for Fintech yet.