{"type":"post","id":"together-compiler-continuous-compilation-for-ai-inference-890880","slug":"together-compiler-continuous-compilation-for-ai-inference-890880","title":"Together Compiler: Continuous Compilation for AI inference","description":"Together AI describes Together Compiler (TKCC), a system that continually searches for inference optimizations tailored to changing models, hardware, and traffic.","retrieval_nugget":"Together AI describes Together Compiler (TKCC), a system that continually searches for inference optimizations tailored to changing models, hardware, and traffic. The company reports up to 2.6× higher tokens per second on selected open-model workloads, highlighting a shift toward treating optimization as a repeated, workload-specific process rather than a one-time tuning exercise.","published_at":"2026-11-04","updated_at":"2026-11-07","record_date":"2026-11-04","date_kind":"published_at","topics":["inference"],"entities":["Together AI","Together Compiler","MiniMax M3.1-Flash","DeepSeek v4.1-Flash","SemiAnalysis","InferenceX","vLLM","SGLang"],"related_patterns":["harness-architecture-outlives-model-choice"],"novelty_label":"new_fact","what_changed":"Together AI introduces its own continuous compilation product for inference; the prior covers a separate research-preview compiler/runtime for planning agent programs.","prior_refs":["https://newruntime.com/posts/agentforge-research-preview-open-source-rust-compiler-and-40ab5b/"],"source_url":"https://www.together.ai/blog/together-compiler-continual-compilation-for-ai-inference","source_urls":["https://www.together.ai/blog/together-compiler-continual-compilation-for-ai-inference"],"schema_version":"newruntime-agent-readable-v0.2","stable_id":"post:together-compiler-continuous-compilation-for-ai-inference-890880","status":"published","visuals":[],"editor":{"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]},"routes":{"html":"https://newruntime.com/posts/together-compiler-continuous-compilation-for-ai-inference-890880/","markdown":"https://newruntime.com/posts/together-compiler-continuous-compilation-for-ai-inference-890880.md","json":"https://newruntime.com/posts/together-compiler-continuous-compilation-for-ai-inference-890880.json"},"markdown_url":"https://newruntime.com/posts/together-compiler-continuous-compilation-for-ai-inference-890880.md","next_reads":[{"type":"topic","path":"/topics/inference/","url":"https://newruntime.com/topics/inference/","title":"Inference","reason":"Explore the inference topic hub.","media_type":"text/html","markdown_url":"https://newruntime.com/topics/inference.md","json_url":"https://newruntime.com/topics/inference.json"},{"type":"collection","path":"/posts/","url":"https://newruntime.com/posts/","title":"Field Notes","reason":"Open the latest editorial analysis.","media_type":"text/html","json_url":"https://newruntime.com/posts.json"},{"type":"collection","path":"/sources/","url":"https://newruntime.com/sources/","title":"Sources","reason":"See the original sources this archive cites.","media_type":"text/html"}]}