Issues

• Using AnyLLM as a provider — cost-efficient Chinese models + OptiLLM techniques (bon/moa/mcts on DeepSeek, Qwen, Kimi)

  algorithmicsuperintelligence/optillm#327 · Filter by author JetragonRider opened on Aug 17, 2026

• proposal: Integration or tracer hook for Agent-Devtools / observability

  algorithmicsuperintelligence/optillm#325 · Filter by author Jacopos311 opened on Aug 15, 2026

• readurls approach fetches attacker-named URLs with no SSRF guard (metadata/internal reachable on hosted proxy)

  algorithmicsuperintelligence/optillm#323 · Filter by author elfrost opened on Jul 18, 2026
  2

• [Feature] Anthropic-compatible API

  enhancementgood first issuehelp wanted
  algorithmicsuperintelligence/optillm#308 · Filter by author qdrddr opened on May 12, 2026
  2

  

• Security: RCE via prompt injection in z3_solver — exec() on LLM-generated code

  algorithmicsuperintelligence/optillm#303 · Filter by author CrepuscularIRIS opened on Apr 11, 2026

• Security: Unauthenticated RCE via executecode plugin — unsandboxed Python execution

  algorithmicsuperintelligence/optillm#302 · Filter by author CrepuscularIRIS opened on Apr 11, 2026

• Proposal: Implement RLM

  enhancement
  algorithmicsuperintelligence/optillm#293 · Filter by author kkontosis opened on Jan 21, 2026

• json response error

  algorithmicsuperintelligence/optillm#292 · Filter by author cometta opened on Jan 21, 2026
  1

• rate limit

  algorithmicsuperintelligence/optillm#291 · Filter by author cometta opened on Jan 21, 2026
  3

• Failed to setup Chrome driver

  algorithmicsuperintelligence/optillm#281 · Filter by author cometta opened on Nov 21, 2025
  3

• Use linters in your project please

  algorithmicsuperintelligence/optillm#275 · Filter by author nnikolov3 opened on Nov 18, 2025
  1

• [Feature Request] Implement Streaming Hint Injection with RL-Trained Retriever

  help wanted
  algorithmicsuperintelligence/optillm#206 · Filter by author codelion opened on Jul 3, 2025
  1

  

• Proposal: Implement CoRT (Code-integrated Reasoning within Thinking) Approach

  enhancementgood first issuehelp wanted
  algorithmicsuperintelligence/optillm#199 · Filter by author codelion opened on Jun 13, 2025

• SafeCoT: Monitoring and Ensuring Chain-of-Thought Faithfulness in Inference-Time Compute

  enhancementgood first issuehelp wanted
  algorithmicsuperintelligence/optillm#198 · Filter by author codelion opened on Jun 11, 2025
  3

• optillm supports logits or logprobs in the API

  documentation
  algorithmicsuperintelligence/optillm#182 · Filter by author codelion opened on Apr 28, 2025
  1

• Add the legacy /completions endpoint

  algorithmicsuperintelligence/optillm#170 · Filter by author codelion opened on Mar 6, 2025

• Small usability improvements

  enhancement
  algorithmicsuperintelligence/optillm#118 · Filter by author av opened on Dec 29, 2024
  1

• Any interest in adding prolog for reasoning problems?

  enhancement
  algorithmicsuperintelligence/optillm#116 · Filter by author ohpauleez opened on Dec 26, 2024
  1

• Add a lighting template for running optillm

  enhancementgood first issue
  algorithmicsuperintelligence/optillm#56 · Filter by author codelion opened on Oct 5, 2024
  1