tabbyapi
Here are 19 public repositories matching this topic...
Call LLMs and VLMs from ComfyUI with any OpenAI compatible chat completions API: local servers (llama.cpp, vLLM, SGLang, Ollama, LM Studio, tabbyAPI, Unsloth Studio…) or cloud providers. Text and image inputs, multi-turn history with memory loops, and model unloading for local servers.
-
Updated
Nov 17, 2026 - Python
Qwen3.8-Flash-Next (turboderp's 2.68bpw EXL3) on 2× RTX 5090, ExLlamaV3 + TabbyAPI: 262k context, 819k-token 8-bit KV pool, 8 slots, MTP + prompt lookup. 1 stream: 239-341 t/s by content with thinking off (311 code), 249 with thinking on; 795 t/s at 8 streams; ~12.4k t/s prefill. Launcher, overlays, probes, dated measurements.
-
Updated
Nov 16, 2026 - Python
Self-hosted AI chat interface with RAG, long-term memory, and admin controls. Works with TabbyAPI, Ollama, vLLM, and any OpenAI-compatible API.
-
Updated
Nov 13, 2026 - Python
GLM-5.3-Flash (321B MoE, 2.05bpw EXL3) on 2× RTX 5090 + 64 GB DDR5, ExLlamaV3 + TabbyAPI. CPU-offloaded experts with runtime hot-expert swaps, exact output: 60 t/s at 1 stream, 110 t/s at 4.
-
Updated
Oct 10, 2026 - Python
Containerized private AI lab: TabbyAPI EXL3 + SillyTavern + Open WebUI + Ollama + SearXNG
-
Updated
Aug 25, 2026 - Jinja
Serve Qwen3.8-Flash-Next (uncensored EXL3) on one NVIDIA DGX Spark: one Docker image, OpenAI-compatible endpoint, MTP speculative decoding, 262k context. ~80 tok/s.
-
Updated
Oct 3, 2026 - Python
Console web locale pour piloter ses LLM : llama.cpp, Strata, Edge0 et TabbyAPI. Chargement a la demande, suivi de la VRAM, MTP et LoRA activables, harness (opencode, Codex, pi, Continue, OpenClaw) synchronises, image/musique/video via MCP. Node.js sans dependance.
-
Updated
Nov 17, 2026 - JavaScript
Reproducible ExLlamaV3 recipe and optimization log for Qwen3.8-Flash-Next on 3x RTX 5070 Ti 16GB
-
Updated
Sep 29, 2026 - Python
A measurement bench for local LLM inference on one RTX 3090. One variable at a time, raw data published, limits declared. Every number comes with the command that produced it.
-
Updated
Oct 31, 2026 - Python
LLM powered chatbot for discussing the Reformed Christian faith (with a lean towards Presbyterianism).
-
Updated
Jun 7, 2025 - Python
LLM powered chatbot for discussing the Reformed Christian faith (with a lean towards Presbyterianism).
-
Updated
Oct 13, 2026 - Python
TabbyAPI fork exposing ExLlamaV3 equal autosplit: same free VRAM on every GPU, per-GPU margins, automatic context with retry, and a load-time stress test. Requires Danielcurly/exllamav3.
-
Updated
Oct 13, 2026 - Python
Add this topic to your repo
To associate your repository with the tabbyapi topic, visit your repo's landing page and select "manage topics."