TNSOR Protocol
All versions of TNSOR Agent for Windows x64. Download any version directly or install via the in-app auto-updater.
Version
Released
Notes
Download
20 Jun 2026
v2.79.0 - Blender pipeline fixes: relay handler, signed URL uploads, RLS bypass
19 Jun 2026
v2.78.1 - Blender Bootstrapper 403 Fix - **Spoof browser User-Agent** in `blender_bootstrapper.py` download request to bypass `download.ble…
19 Jun 2026
v2.78.0 - Headless Parameterized Blender Render Engine New Features - **Blender Bootstrapper**: Auto-downloads and caches portable Blender …
19 Jun 2026
v2.77.0 - JWT Expiry Resilience - **Upload fallback**: When JWT returns 401/403, automatically retry with anon key so generated images are …
19 Jun 2026
Root Cause Module-level `_load_model_config()` printed config logs to stdout before `check_modules` could emit its JSON result. The Tauri Ru…
19 Jun 2026
Fixes - **VRAM Gatekeeper fallback**: When nvidia-smi is unavailable, engine now probes via portable python + torch before returning 0 GB - …
31 May 2026
v2.75.0 — Rollback: Restore Clean v2.73.0 LLM-Only Codebase Reverts v2.74.0 (Blender engine) completely. All Blender-related code, workers,…
30 May 2026
TNSOR Agent v2.74.0 — Distributed Blender Render Engine New: Full-Stack Blender 3D Rendering Pipeline - **Automated Bootstrapper**: Portabl…
30 May 2026
v2.73.0 — Decoupled Hardware Registry (Dynamic Runtime Hydration) All hardcoded HuggingFace model strings removed from the engine and front…
30 May 2026
v2.72.0 — Host UI: Private LLM Card Sync with Image Gen What changed The Private LLM compute module card on the Host Dashboard now mirrors …
30 May 2026
v2.71.0 — Dynamic LLM Routing, Chat Templates, Model Selector UI Phase 1 — Worker: Dynamic Model Loading + Chat Templates - Worker accepts …
30 May 2026
v2.70.0 — Persistent LLM Daemon (Warm REPL) Eliminates the 8-second cold-start per message. Model loads once; subsequent prompts dispatched…
30 May 2026
v2.69.0 — UTF-8 Subprocess Encoding Patch **Fixes:** UnicodeEncodeError: charmap codec crash when LLM generates non-ASCII tokens on Windows…
30 May 2026
v2.68.0 — Private LLM Streaming Pipeline Phase 1 — OS-Level I/O Hardening - stdin=DEVNULL, stderr=STDOUT, CREATE_NO_WINDOW on all LLM subpr…
29 May 2026
feat: v2.65.0 implement VRAM gatekeeper via nvidia-smi, dynamic workload registry, and UI model selection toggles
29 May 2026
feat: v2.64.0 inject physical CUDA telemetry and VRAM allocation checks into inference worker without altering architecture - Prints `[AI W…
29 May 2026
fix: v2.63.0 resolve missing file deadlock during authentication, shift debug logs pre-auth, and preserve subprocess I/O architecture - Pha…
29 May 2026
fix: v2.62.0 add transformers to image_generation pip requirements to resolve diffusers text_encoder dependency
29 May 2026
fix: v2.61.0 resolve Windows GUI stdin deadlock, prevent CPU busy-wait loops, and disable tqdm carriage returns
29 May 2026
fix: v2.60.0 ultimate architecture patch - enforce OS-level stderr merging, eliminate pipe buffer deadlocks, and implement safe non-blocking…
26 May 2026
fix: v2.59.0 restore worker script early prints, unbuffered stdout, and tokenizer safety to fix silent first-boot downloads
26 May 2026
chore: v2.58.2 fix path recon race - write to file + re-emit after WS connect
26 May 2026
chore: v2.58.1 inject filesystem path reconnaissance logs to debug silent pathing mismatch
26 May 2026
v2.58.0 - DO OR DIE: Ruthless JIT Workload Interception + Post-Mount Reality Check Why v2.57.0 Failed Two independent failures: 1. **Tauri …
26 May 2026
v2.57.0 - DO OR DIE: Boot-Time UI State Sync + Total Thread Isolation Root Cause of v2.56.0 Failure The frontend calls `check_modules` exac…
26 May 2026
v2.56.0 - Hardening Patch: Strict AI Core Health Checks Problem The engine was hallucinating its health state. After manually deleting `tns…
25 May 2026
v2.55.1 - Fix Inference Worker Subprocess Freeze Problem Image generation froze permanently at [AI] SD-Turbo worker PID=XXXXX - waiting for…
25 May 2026
v2.55.0 - Fix Image Generation Root Causes Fixed - **Stale busy flag**: `gen_state["busy"]` was never reset on WebSocket reconnect. Any cra…
25 May 2026
v2.54.0 — Hard revert: restore v2.51.0 ThreadPoolExecutor architecture Critical fix - **Image generation fully restored**: Removed `asyncio…
24 May 2026
v2.53.0 — Emergency fix: revert blocking I/O deadlock from v2.52.0 Critical fix - **Deadlock eliminated**: v2.52.0 introduced `text=True + …
All releases are signed with the TNSOR minisign keypair and verified by the in-app auto-updater.