algorithmicsuperintelligence/openevolve

Issues

• [Question]LLM prompt evolution optimisation with feedback

  #315 opened 3 days ago
  1

• ERROR - Error in novelty LLM check: asyncio.run() cannot be called from a running event loop

  #313 opened 5 days ago
  4

• Optimizing assembly code

  #312 opened 6 days ago
  2

• Include competition between evolved populations

  #311 opened 7 days ago
  1

• Suggestion: replace code diversity measure with levenshtien distance

  #309 opened 13 days ago
  1

• is that correct for next_island called before increment_island_generation?

  #307 opened 16 days ago
  1

• Question: OpenEvolve score range

  #306 opened 16 days ago
  0

• Bug: rust program fails to compile due to type errors in example

  #304 opened 19 days ago
  0

• Bug: async function evaluate in Rust adaptive sort example

  #303 opened 19 days ago
  0

• "Performs well on" key_features added to prompts

  #299 opened 24 days ago
  1

• Incorrectly recognizing OpenAI models based on API url

  #298 opened 23 days ago
  1

• Evaluator timeouts do not seem to work, system still hangs

  #297 opened a month ago
  1

• Multiple Openevolve run calls in parallel result in clobbered loggers

  #295 opened a month ago
  2

• How to handle early stopping patience with failure cases

  #294 opened a month ago
  3

• Can Different Islands Be Configured With Different Prompts to Support Diverse Evolutionary Directions?

  #293 opened a month ago
  1

• Feature Request: Logger prefix

  #290 opened a month ago
  1

• Bug: other islands do not increment their generations

  #289 opened a month ago
  12

• openevolve does not correctly clean up when there are failures that lead to retries

  #288 opened a month ago
  3

• The evaluator for circle packing can be hacked

  #285 opened a month ago
  2

• Explanation / Documentation Request: use_llm_feedback, cascade_evaluation, and use_template_stochasticity

  #284 opened a month ago
  4

• Support Hydra/OmegaConf composition in load_config

  #282 opened a month ago
  0

• Best program is the same as the initial one

  #281 opened a month ago
  1

• Suggestion: Use AsyncOpenAI instead of the default OpenAI package

  #278 opened a month ago
  0

• Implement efficient sampling techniques

  #277 opened a month ago
  3

• Tree search and/or agentic behavior

  #276 opened 2 months ago
  4

• Error when optimizing .cpp program

  #271 opened 2 months ago
  0

• attention_optimization initial_program.py has no evolve block

  #270 opened 2 months ago
  2

• [Question] Process for creating system prompt.

  #269 opened 2 months ago
  3

• BUG: random seed is reset in PromptSampler resulting in nondeterministic runs

  #267 opened 2 months ago
  0

• Quickstart yields "No LLM models configured."

  #264 opened 2 months ago
  1

• Bug: reasoning_effort is not properly ingested

  #259 opened 2 months ago
  0

• LLM Reflection/GEPA

  #258 opened 2 months ago
  3

• Request: stopping at a threshold score

  #257 opened 2 months ago
  5

• Request: example config that just runs a single step / iteration

  #255 opened 2 months ago
  3

• README is incorrect

  #253 opened 2 months ago
  1

• A

  #252 opened 2 months ago
  0

• Missing comma

  #247 opened 2 months ago
  0

• i need help , v0.2.8 ,Frequently occurs, A process in the process pool was terminated abruptly while the future was running or pending.

  #246 opened 2 months ago
  15

• Command-line overrides doesn't work for CLI

  #245 opened 2 months ago
  1

• System messages for llm prompt optimization

  #243 opened 2 months ago
  10

• Handling prompts with different placeholder names

  #242 opened 2 months ago
  1

• No valid diffs found in response and programs not evolving

  #241 opened 2 months ago
  2

• openevolve checkpoint cannot save the proper number of evolved programs

  #239 opened 2 months ago
  4

• a minor patch to support gpt-5

  #225 opened 3 months ago
  1

• feature_dimensions is not defined in sampler.py _format_inspirations_section

  #223 opened 3 months ago
  0

• Bug: Programs' parents are not in the corresponding islands

  #222 opened 3 months ago
  0

• Use combined_score in PromptSampler when available

  #220 opened 3 months ago
  1

• Can the LLM engine be replaced by litellm?

  #219 opened 3 months ago
  3

• Incorrect or Confusing Top and Previous Programs

  #214 opened 3 months ago
  2

• Database.py inconsistent use of combined_score and safe_numeric_average

  #213 opened 3 months ago
  1