Standalone local OpenAI-compatible gateway for OpenFang coding, agent workflows, high-context coding lanes on the paired 32 GB and 16 GB V100s, pressure-aware chat routing across Qwen3.6, North Mini Code, and Qwen3-8B, V100 12GB primary memory retrieval with gateway cache and upstream proof, P40 utility chat, P100 emergency embedding failover and reranking, and RTX 2080 Ti media work. It is intentionally separate from the stopped RassyGPT Runtipi app.
| Model | Kind | Backend | Model Used | Hardware | Description |
|---|---|---|---|---|---|
rassy-codex | chat | codex | qwen3.6:35b | 2x Tesla V100-PCIE-32GB | Qwen3.6 35B high-context coding and OpenFang operator lane on the paired 32 GB V100 cards. Features: agentic coding, thinking preservation, tool use, vision-capable upstream |
rassy-codex-lite | chat | codex_lite | qwen3.6:35b | 2x Tesla V100-PCIE-32GB lower-context coding lane | Lower-context Qwen3.6 coding lane on the paired 32 GB V100 cards for normal-sized agent work. Features: lower-context coding, isolated normal-sized coding queue, faster agent loops |
rassy-general | chat | general | north-mini-code-1.0:q4_K_M | 2x Tesla V100 16GB | North Mini Code 1.0 high-context agentic software-engineering lane on the paired 16 GB V100 cards. Features: agentic coding, interleaved thinking, tool use, long-horizon repo work |
rassy-fast | chat | fast | unsloth/Qwen3-8B-GGUF / Qwen3-8B-Q4_K_M.gguf | Tesla P40 24GB utility chat lane | Qwen3-8B utility chat lane on the P40 for summaries, routing, compaction, and former worker aliases. Features: utility summaries, routing, compaction, compatibility worker aliases |
rassy-embed | embeddings | embed | qwen3-embedding:8b | Tesla V100-PCIE-12GB primary embedding lane | Ollama Qwen3-Embedding-8B document and memory embedding lane on the V100 12GB, with P100 standby after primary failure or bounded primary stall. Features: 4096-dim Ollama document embeddings, OpenFang memory retrieval, P100 emergency failover after failure or bounded stall |
rassy-embed-query | embeddings | embed_query | qwen3-embedding:8b | Tesla V100-PCIE-12GB shared query/document embeddings | Query embeddings share the V100 12GB embedding service with RassyCodex retrieval instructions and P100 standby after primary failure, bounded primary stall, or query-only overflow pressure. Features: query-prefixed embeddings, same V100 12GB service as document embeddings, P100 emergency failover after failure or bounded stall |
rassy-embed-p100 | embeddings | embed_p100 | qwen3-embedding:8b | Tesla P100 16GB emergency embedding failover | Conservative Qwen3-Embedding-8B standby on the P100; explicit use or automatic failover only. Features: 4096-dim Ollama embeddings, emergency failover, P100 co-resident with rerank |
rassy-rerank | rerank | rerank | ExpedientFalcon/Qwen3-Reranker-4B-Q4_K_M-GGUF / qwen3-reranker-4b-q4_k_m.gguf | Tesla P100 16GB | Qwen3-Reranker-4B Q4 lane on the P100 for retrieval self-tests, with Qwen embedding cosine fallback. Features: retrieval reranking, cosine fallback |
rassy-image | images | image | flux.2-klein-4b | RTX 2080 Ti 11GB media lane | LocalAI image/photo generation lane on the RTX 2080 Ti. Features: local photo/image generation, RTX 2080 Ti media lane |
rassy-audio | audio | audio | whisper-large-turbo-q8_0 / qwen3-tts-cpp | RTX 2080 Ti 11GB media lane | LocalAI STT/TTS lane on the RTX 2080 Ti. Features: local STT, local TTS, RTX 2080 Ti media lane |