The Deploy Log | Tools | Free, no signup
LLM inference engines
7 LLM inference engines in The Deploy Log, side by side: 41 versions between them, 14 in the last 30 days. Ollama shipped the most of them, 3, then vLLM with 2. Every item sourced to the publisher's own page.
| Tool | Company | Newest | Its day | Last 30 days | A new one |
|---|---|---|---|---|---|
| Ollama | Ollama | v0.40.17 | Undated until its day | 3 | every 7 days |
| vLLM | vLLM | v0.34.0 | 2 | every 9 days | |
| MLX-VLM | MLX-VLM | v0.8.0 | 2 | every 12 days | |
| llama.cpp | llama.cpp | v1.0.0 | 2 | every 11 days | |
| Dynamo | NVIDIA | v1.6.2 | 2 | every 13 days | |
| ONNX Runtime | Microsoft | v1.32.0 | 2 | every 21 days | |
| KoboldCpp | KoboldCpp | 1.124.1 | 1 | every 17.5 days |
The newest releases across the 7
- MLX-VLM v0.8.0
- vLLM v0.34.0
- llama.cpp v1.0.0
- ONNX Runtime v1.32.0
- vLLM v0.34.0rc1
- Ollama v0.40.16
- Dynamo v1.6.2
- llama.cpp v0.9.0
- MLX-VLM v0.7.12
- Ollama v0.40.14
- KoboldCpp 1.124.1
- Ollama v0.40.13
- ONNX Runtime v1.30.1
- Dynamo v1.6.0
- MLX-VLM v0.7.10
- Ollama v0.40.8
- MLX-VLM v0.7.9
- KoboldCpp 1.123.1
- llama.cpp v0.7.0
- MLX-VLM v0.7.8
One line every time AI actually ships, around the clock, free on Threads and Bluesky. Every one is in The Deploy Log with its source.