RigBase
Local LLM hardware benchmarks with measured tokens per second from real home AI rigs
- Upvotes
- 0
- DR
- 0 no change since last check
- Followers
- 0
Database of real home AI rig builds showing measured LLM performance, GPU specifications, power draw and costs across 172 community-sourced configurations.
About RigBase
RigBase is a benchmark database aggregating real-world home AI builds and their measured performance running local large language models. The platform collects data from 172 actual AI rigs spanning various hardware configurations including GPUs, Apple Macs, Strix Halo systems and DGX Spark units.
For each build, RigBase provides detailed specifications including component parts, total cost, power consumption, and actual measured token generation speeds. The database covers popular current models including Qwen3.8 27B, Gemma 4 31B, gpt-oss 120B, and others, with every performance metric linked to its original public source.
The platform aggregates data from established benchmarking sources such as llama-bench, llama.cpp discussions, and ai-benchmarks. Users can search by GPU type, VRAM size, budget constraints, and target model to find real-world reference builds. The full dataset is available for download under a Creative Commons BY 4.0 license.
- ai
- open-source
- performance
- local llm
- hardware
- gpu benchmark
Features worth knowing
- 172 real home AI builds with measured performance data
- Performance metrics for 31 current LLM models
- Search by GPU, VRAM, budget and target model
- VRAM and KV cache calculator
- All results linked to original public sources
- Comparison guides and hardware recommendations
Pricing
Pricing model: Free
RigBase metrics
| Metric | Value |
|---|---|
| Domain Rating | 0 no change since last check |
| Trust Flow | 0 no change since last check |
| Citation Flow | 20 no change since last check |
| Referring domains | 2 no change since last check |
| External backlinks | 9 no change since last check |
Ahrefs Domain Rating scanned 28 Oct 2026, updated 28 Oct 2026. Majestic scanned 28 Oct 2026, updated 28 Oct 2026. Link stats via Majestic. Trust Flow, Citation Flow and Topical Trust Flow are trademarks of Majestic-12 Ltd.
RigBase domain facts
Reveal the domain facts
| Domain age | 9 days |
|---|---|
| Registered | |
| Expires | |
| Registrar | NameCheap, Inc. |
| DNSSEC | Not signed |
Compare and explore
FAQ
- What LLM models are benchmarked in RigBase?
- RigBase covers 31 current models including Qwen3.8 27B, Gemma 4 31B, gpt-oss 120B, Qwen3.6 35B-A3B, gpt-oss 20B, Gemma 4 26B-A4B, Qwen3.8 Flash-Next, GLM-5.3 Flash, and DeepSeek V4.1 Flash, with 54 legacy models also included.
- What types of hardware are included in the benchmarks?
- RigBase includes real builds with GPUs (RTX 5090, RTX 4090, RTX 3090, Radeon AI Pro), Apple Macs (M1 Ultra, M3 Ultra, M5 Max), Strix Halo systems, and DGX Spark/Nvidia GB10 configurations.
- How can I download the RigBase dataset?
- The full RigBase dataset is available as free open data under Creative Commons BY 4.0 license in CSV, JSON, and HTML formats.
- Are the benchmark results verified or marketing numbers?
- All results are real measured data aggregated from public benchmarks like llama-bench and llama.cpp discussions, not re-measured by RigBase. Every number links to its original public source.
Find RigBase online
Leave a review for RigBase
Nobody has reviewed RigBase yet. Start things off.
A review is always free and can never be paid for. Go to the reviews and discussion
Reviews and chat · Product news · Send it on · Run this company? Claim it here · Report it
RigBase has a free listing. The owner can unlock the full profile with a paid launch for $5:
- Founder story (locked)
- Milestones (locked)
- Demo video (locked)
- Dofollow website link (locked)
Alternatives to RigBase
Tensordyne
Next-generation AI inference systems for speed and efficiency
AI inference & GPU cloud0 votes. Log in to add your vote for TensordyneJua
The most accurate AI weather forecasting model for energy markets
AI inference & GPU cloud0 votes. Log in to add your vote for JuaMultiverse Computing
Efficient and secure AI with tailored solutions reducing compute costs across cloud, data centers and edge environments
AI inference & GPU cloud0 votes. Log in to add your vote for Multiverse ComputingScale
Reliable AI systems for critical decisions
AI inference & GPU cloud0 votes. Log in to add your vote for Scale
