https://pypi.org/project/squish-ai/ Recent updates to the Python Package Index for squish-ai en https://pypi.org/project/squish-ai/9.33.8/ Local LLM inference server for Apple Silicon. Block-level paged KV cache for long-context workloads. 5.4× faster end-to-end on 4K-token prompts vs Ollama, less RAM, INT3 support for Qwen3. OpenAI-compatible API. Tue, 09 Jun 2026 16:57:00 GMT https://pypi.org/project/squish-ai/9.33.7/ Local LLM inference server for Apple Silicon. Block-level paged KV cache for long-context workloads. 5.4× faster end-to-end on 4K-token prompts vs Ollama, less RAM, INT3 support for Qwen3. OpenAI-compatible API. Mon, 08 Jun 2026 15:50:50 GMT https://pypi.org/project/squish-ai/9.33.6/ Local LLM inference server for Apple Silicon. Block-level paged KV cache for long-context workloads. 5.4× faster end-to-end on 4K-token prompts vs Ollama, less RAM, INT3 support for Qwen3. OpenAI-compatible API. Sat, 06 Jun 2026 02:47:13 GMT https://pypi.org/project/squish-ai/9.33.5/ Local LLM inference server for Apple Silicon. Block-level paged KV cache for long-context workloads. 5.4× faster end-to-end on 4K-token prompts vs Ollama, less RAM, INT3 support for Qwen3. OpenAI-compatible API. Fri, 05 Jun 2026 17:30:38 GMT https://pypi.org/project/squish-ai/9.33.4/ Local LLM inference server for Apple Silicon. Block-level paged KV cache for long-context workloads. 5.4× faster end-to-end on 4K-token prompts vs Ollama, less RAM, INT3 support for Qwen3. OpenAI-compatible API. Thu, 04 Jun 2026 18:10:20 GMT https://pypi.org/project/squish-ai/9.33.2/ Local LLM inference server for Apple Silicon. Block-level paged KV cache for long-context workloads. 5.4× faster end-to-end on 4K-token prompts vs Ollama, less RAM, INT3 support for Qwen3. OpenAI-compatible API. Wed, 03 Jun 2026 17:48:27 GMT https://pypi.org/project/squish-ai/9.33.1/ Local LLM inference server for Apple Silicon. Block-level paged KV cache for long-context workloads. 5.4× faster end-to-end on 4K-token prompts vs Ollama, less RAM, INT3 support for Qwen3. OpenAI-compatible API. Wed, 03 Jun 2026 17:31:07 GMT https://pypi.org/project/squish-ai/9.33.0/ Local LLM inference server for Apple Silicon. Block-level paged KV cache for long-context workloads. 5.4× faster end-to-end on 4K-token prompts vs Ollama, less RAM, INT3 support for Qwen3. OpenAI-compatible API. Wed, 03 Jun 2026 17:24:45 GMT https://pypi.org/project/squish-ai/9.32.0/ Local LLM inference server for Apple Silicon. Block-level paged KV cache for long-context workloads. 5.4× faster end-to-end on 4K-token prompts vs Ollama, less RAM, INT3 support for Qwen3. OpenAI-compatible API. Tue, 02 Jun 2026 18:41:36 GMT