C++ · MIT
Indras-Mirror/Xolotl
Xolotl (formerly Strata-DSV4.1): run DeepSeek-V4.1-Flash, all 552B, as a local coding agent on one gaming PC (RTX 4090 + 90 GB RAM) at up to ~27 tok/s, 512K context, October 1017 builds included. Plus GLM-5.3-Flash and Qwen3.8. MoE experts tiered across VRAM, pinned RAM and NVMe; OpenAI + Anthropic API.
★ 460
Stars
58 forks · 15 open issues
+8.7/day
Growth · measured
Stars counted between two of our own snapshots roughly a week apart. The repository page gives the exact span.
41.5
Momentum / 100
see the breakdown below
32d ago
Created
last push 3d ago
Star history
How this score was built
| Component | Weight | Earned | Why |
|---|---|---|---|
| Velocity | 55 | 18.9 | +8.7/day, log-scaled against a 750/day ceiling |
| Acceleration | 15 | 0.0 | last week ran 0.41× the monthly pace |
| Freshness | 15 | 14.9 | last push 3d ago; decays to zero at 180 days |
| Scale | 15 | 7.7 | 460 stars, log-scaled against 150k — stops a 3-star repo topping the board |
| Momentum | 100 | 41.5 | sum of the above |
Detail
| Topics | abliterationcudadeepseekdeepseek-v4deepseek-v4-1ggufglmglm-5llm-inferencelocal-llmmixture-of-expertsmoeoffloadingproject-mayaquantizationqwenrtx-4090stratauncensored |
| Matched on | LLM & Inference (llm-inference) |
| Short window | +61 stars over 7 days = 8.7/day |
| Long window | +385 stars over 18 days = 21.4/day |
| Lifetime average | 15.46 stars/day over 30 days |
| First seen here | 2026-10-18 |
| Last read | 2d ago |