Marco-o1
An early open reasoning model aimed at open-ended problems.
About
Marco-o1 is an open reasoning model from Alibaba International's MarcoPolo team, one of the earlier efforts to push chain-of-thought style reasoning toward open-ended, real-world tasks rather than just math puzzles. It ships with inference scripts including vLLM support and is Apache-2.0 licensed. The project is small and research-grade, and the team is upfront that it cannot guarantee the training data is fully clean. It has largely been eclipsed by newer open reasoning models, but it remains a usable starting point for reasoning experiments.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Paper reports improvements on open-ended reasoning evaluations (arXiv 2411.14405, authors).
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 6/10
Scored 6/10 (verified tier, metrics-v1). Evidence: Small research project from Alibaba International's MarcoPolo team (paper Nov 2024); Apache-2.0 licensed; ships vLLM inference scripts; Low community activity; largely eclipsed by newer open reasoning models; Team discloses it cannot guarantee training-data cleanliness. License: Apache-2.0 per the project README..
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 5.0/10
- Maintenance · recency, releases, issue load
- 4.0/10
- Openness · license + docs
- 8.0/10
- Manual review · sanity read + red-flag check
- 6.0/10
- · Small research project from Alibaba International's MarcoPolo team (paper Nov 2024)
- · Apache-2.0 licensed; ships vLLM inference scripts
- · Low community activity; largely eclipsed by newer open reasoning models
- · Team discloses it cannot guarantee training-data cleanliness