fixes

Author: codelion

CLAUDE.md

@@ -4,7 +4,7 @@ This file provides guidance to Claude Code (claude.ai/code) when working with co
 
 ## Project Overview
 
-OptILLM is an OpenAI API compatible optimizing inference proxy that implements state-of-the-art techniques to improve accuracy and performance of LLMs. It focuses on reasoning improvements for coding, logical, and mathematical queries through inference-time compute optimization.
+OptiLLM is an OpenAI API compatible optimizing inference proxy that implements state-of-the-art techniques to improve accuracy and performance of LLMs. It focuses on reasoning improvements for coding, logical, and mathematical queries through inference-time compute optimization.
 
 ## Core Architecture
 
@@ -95,7 +95,7 @@ python scripts/eval_arena_hard_auto_rtc.py
 # FRAMES benchmark
 python scripts/eval_frames_benchmark.py
 
-# OptILLM benchmark generation/evaluation
+# OptiLLM benchmark generation/evaluation
 python scripts/gen_optillmbench.py
 python scripts/eval_optillmbench.py
 ```

README.md

@@ -641,7 +641,7 @@ called patchflows. We saw huge performance gains across all the supported patchf
 
 ## Testing
 
-OptILLM includes a comprehensive test suite to ensure reliability and compatibility.
+OptiLLM includes a comprehensive test suite to ensure reliability and compatibility.
 
 ### Running Tests
 

optillm/deepconf/README.md

@@ -14,7 +14,7 @@ Based on the paper "Deep Think with Confidence" by Fu et al. (2024), DeepConf im
 
 ## Features
 
-- ✅ **Local models only** - Works with OptILLM's local inference engine
+- ✅ **Local models only** - Works with OptiLLM's local inference engine
 - ✅ **Two variants**: `low` (aggressive, top 10%) and `high` (conservative, top 90%)
 - ✅ **Configurable parameters** for different use cases
 - ✅ **Early termination** to reduce token usage by 50-70%
@@ -24,7 +24,7 @@ Based on the paper "Deep Think with Confidence" by Fu et al. (2024), DeepConf im
 
 ### Basic Usage
 
-Set up OptILLM for local inference:
+Set up OptiLLM for local inference:
 
 ```bash
 export OPTILLM_API_KEY=optillm
@@ -120,7 +120,7 @@ DeepConf typically achieves:
 ## Requirements
 
 - Local model inference (PyTorch)
-- OptILLM with `OPTILLM_API_KEY=optillm`
+- OptiLLM with `OPTILLM_API_KEY=optillm`
 - Compatible with transformer models that provide logits access
 
 ## Limitations

optillm/plugins/deep_research/README.md

@@ -22,7 +22,7 @@ deep_research/
 ├── research_engine.py       # Core TTD-DR implementation
 └── README.md               # This documentation
 
-../deep_research_plugin.py  # OptILLM plugin interface
+../deep_research_plugin.py  # OptiLLM plugin interface
 ```
 
 ### Key Components
@@ -41,7 +41,7 @@ The core implementation of the TTD-DR algorithm with the following key methods:
 
 #### 2. Plugin Interface (`deep_research_plugin.py`)
 
-Minimal interface that integrates with OptILLM's plugin system:
+Minimal interface that integrates with OptiLLM's plugin system:
 
 ```python
 def run(system_prompt: str, initial_query: str, client, model: str, request_config: Optional[Dict] = None) -> Tuple[str, int]
@@ -101,7 +101,7 @@ request_config = {
 
 ## Dependencies
 
-The Deep Research plugin requires these OptILLM plugins:
+The Deep Research plugin requires these OptiLLM plugins:
 
 - **`web_search`** - Chrome-based Google search automation
 - **`readurls`** - Content extraction from URLs
@@ -137,7 +137,7 @@ result, tokens = run(
 )
 ```
 
-### With OptILLM Server
+### With OptiLLM Server
 
 ```python
 from openai import OpenAI

optillm/plugins/deep_research/sample_reports/01_evaluate_the_potential_consequences_of_tiktok_bans.md

@@ -197,4 +197,4 @@ By embracing these strategies, companies can mitigate investment risks associate
 [50] TikTok vs. Reels vs. Shorts (A Study by Socialinsider). Available at: https://www.socialinsider.io/blog/tiktok-vs-reels-vs-shorts?ref=blog-cms.socialinsider.io [Accessed: 2025-07-25]
 
 ---
-*Generated using [OptILLM Deep Research](https://github.com/codelion/optillm) with TTD-DR (Test-Time Diffusion Deep Researcher)*
+*Generated using [OptiLLM Deep Research](https://github.com/codelion/optillm) with TTD-DR (Test-Time Diffusion Deep Researcher)*

optillm/plugins/deep_research/sample_reports/02_chart_the_developing_landscape_of_ai_agents_and_co.md

@@ -119,4 +119,4 @@ The landscape of AI agents and their supporting technologies is exceptionally dy
 [12] Reinforcement Learning: Applications in Gaming, Robotics .... Available at: https://www.researchgate.net/publication/390582934_Reinforcement_Learning_Applications_in_Gaming_Robotics_and_Real-World_Decision-Making#:~:text=Applications%20include%20robotic%20manipulation%2C%20autonomous,%2Dworld%20decision%2Dmaking%20processes. [Accessed: 2025-07-25]
 
 ---
-*Generated using [OptILLM Deep Research](https://github.com/codelion/optillm) with TTD-DR (Test-Time Diffusion Deep Researcher)*
+*Generated using [OptiLLM Deep Research](https://github.com/codelion/optillm) with TTD-DR (Test-Time Diffusion Deep Researcher)*

optillm/plugins/deep_research/sample_reports/03_analyze_the_dynamic_terrain_of_unbanked_market_acc.md

@@ -147,4 +147,4 @@ The "Unbanked Market Access" represents a complex yet highly promising domain. S
 [25] The Role of Regulatory Sandboxes in FinTech Innovation. Available at: https://www.mdpi.com/2674-1032/4/2/26 [Accessed: 2025-07-25]
 
 ---
-*Generated using [OptILLM Deep Research](https://github.com/codelion/optillm) with TTD-DR (Test-Time Diffusion Deep Researcher)*
+*Generated using [OptiLLM Deep Research](https://github.com/codelion/optillm) with TTD-DR (Test-Time Diffusion Deep Researcher)*

optillm/plugins/deep_research/sample_reports/04_examine_kkrs_tech_centric_transactions_and_their_a.md

@@ -154,4 +154,4 @@ KKR has established a well-defined and evolving strategy for investing in and ge
 [20] Tech Growth. Available at: https://www.kkr.com/invest/private-equity/tech-growth [Accessed: 2025-07-25]
 
 ---
-*Generated using [OptILLM Deep Research](https://github.com/codelion/optillm) with TTD-DR (Test-Time Diffusion Deep Researcher)*
+*Generated using [OptiLLM Deep Research](https://github.com/codelion/optillm) with TTD-DR (Test-Time Diffusion Deep Researcher)*

optillm/plugins/deep_research/sample_reports/05_break_down_khosla_ventures_portfolio_into_cutting.md

@@ -123,4 +123,4 @@ Khosla Ventures has firmly established itself as a significant force in the vent
 [16] Portfolio. Available at: https://www.khoslaventures.com/portfolio/ [Accessed: 2025-07-25]
 
 ---
-*Generated using [OptILLM Deep Research](https://github.com/codelion/optillm) with TTD-DR (Test-Time Diffusion Deep Researcher)*
+*Generated using [OptiLLM Deep Research](https://github.com/codelion/optillm) with TTD-DR (Test-Time Diffusion Deep Researcher)*

optillm/plugins/deep_research/sample_reports/06_evaluate_the_commercial_rationale_for_participatin.md

@@ -304,4 +304,4 @@ SSRN. (n.d.). [Relevant paper on machine learning for house price prediction]. R
 [52] Many first-time homebuyers are pushing 40 as millennials .... Available at: https://www.nbcnews.com/business/real-estate/many-first-time-homebuyers-are-pushing-40-millennials-wait-vain-better-rcna201786 [Accessed: 2025-07-25]
 
 ---
-*Generated using [OptILLM Deep Research](https://github.com/codelion/optillm) with TTD-DR (Test-Time Diffusion Deep Researcher)*
+*Generated using [OptiLLM Deep Research](https://github.com/codelion/optillm) with TTD-DR (Test-Time Diffusion Deep Researcher)*