A collection of LLM papers, blogs, and projects, with a focus on OpenAI o1 🍓 and reasoning techniques.
6.9k
Stars
370
Forks
27
Open issues
27
Contributors
AI Analysis
A curated collection of research papers, blogs, and projects focused on OpenAI o1/Strawberry and LLM reasoning techniques, including chain-of-thought, MCTS, and reinforcement learning approaches. This is a specialized research aggregation tool best suited for researchers, ML engineers, and practitioners actively exploring reasoning-enhanced language models—not a general-purpose LLM library or production framework.
Inferred from signals mentioned in the README (tests, CI, type safety) — not a review of the actual code.
AI's overall editorial judgment — not an average of the bars above, can weigh other factors too.
Curated reading list tracking the LLM reasoning and OpenAI o1 research wave
Awesome-LLM-Strawberry is a manually curated index of papers, blog posts, talks, and news items focused on LLM reasoning techniques — particularly the methods behind OpenAI's o1/o3 series and related work from DeepSeek, Anthropic, and others. It targets ML researchers, AI engineers, and practitioners trying to follow a fast-moving subfield. With ~6,900 stars and 370 forks, it saw strong early traction after o1's September 2024 launch but has since slowed. Its value is as a structured entry point into inference-time compute and RLVR literature.
Created on September 15, 2024 — within days of OpenAI's o1 public reveal — it rode the initial wave of researcher interest in chain-of-thought and inference-time scaling. The maintainer also appears to be an active reasoning researcher.
Growth was almost entirely event-driven: the o1 announcement triggered a spike in stars in late 2024, with updates through December 2024 covering o3, DeepSeek-R1, and related releases. Stars gained in the last 7 days are 0, indicating growth has fully plateaued as of mid-2026.
adoption not verified — this is a reference list, not deployable software. Indirect signals include 370 forks (suggesting researchers are building on or mirroring it) and citations in blog posts, but no direct production usage evidence exists by definition.
This is a documentation/curation repository, not a software project. It appears to be a structured Markdown file with categorized links — no source code, tooling, or runnable components are present based on the README.
not documented in README
Last push was December 17, 2025 — approximately 6 months before the evaluation date. Given the rapid pace of the LLM reasoning field in 2025-2026, a 6-month gap is significant. The repository appears to have entered low-maintenance or dormant status, though it has not been archived.
ADOPT IF: you need a quick structured entry point into the o1/o3-era reasoning literature from late 2024 and want curated pointers to key papers, blogs, and talks from that period. AVOID IF: you need coverage of reasoning research from 2025 onward, as the list appears unmaintained for ~6 months and the field has moved substantially. MONITOR IF: the maintainer resumes updates — their own research background (REINFORCE++, MoE RL) suggests they have the expertise to produce a high-quality continued resource if they choose to.
Independent dimensions
Mainstream potential
2/10
Technical importance
4/10
Adoption evidence
2/10
- The repository appears to have stopped receiving updates in December 2025, making it increasingly stale in a field that evolves in weeks.
- No programmatic tooling, search, or filtering means discoverability degrades as the list grows long and outdated entries accumulate.
- Single-maintainer dependency: if the author deprioritizes this project, there is no contributor community evident to sustain it.
- The original focus on 'o1 specifically' has been superseded by a much broader reasoning ecosystem (o3, DeepSeek-R2, Gemini thinking models), making the framing partially obsolete.
- Zero stars in the last 7 days suggests organic discovery has effectively ceased, limiting its role as a living community resource.
Likely to remain a useful historical snapshot of the 2024 reasoning research moment but will lose relevance as a current reference as more actively maintained alternatives emerge or expand their scope.
Newsletter
Get analyses like this every Monday
Free weekly digest of the most interesting open-source discoveries.
Languages
No language breakdown available.
Information
- License
- Apache-2.0
- Last updated
- 7mo ago
- Created
- 22mo ago
- Analyzed with
- anthropic/claude-haiku-4-5
Stars over time
No commit data available.
Contributors over time
Top 100 contributors only — repos with more will plateau at 100.
Open issues
No open issues — clean slate.
Open pull requests
No open pull requests.
Top contributors
Contributor data not available yet.
Recent releases
No releases published yet.
Similar repos
TsinghuaC3I/Awesome-RL-for-LRMs
This is a curated survey and collection of papers on reinforcement learning...
Hannibal046/Awesome-LLM
Awesome-LLM is a curated index of Large Language Model resources, including...
WangRongsheng/awesome-LLM-resources
This is a curated index of Large Language Model resources, covering multimodal...
xlite-dev/Awesome-LLM-Inference
A curated collection of academic papers and open-source implementations focused...
| Repository | Stars | Week Δ | Language | Score | Updated |
|---|---|---|---|---|---|
|
hijkzzz/Awesome-LLM-Strawberry
this
|
6.9k | — | — | 7/10 | 7mo ago |
|
TsinghuaC3I/Awesome-RL-for-LRMs
|
2.5k | — | TeX | 8/10 | 8mo ago |
|
Hannibal046/Awesome-LLM
|
27.1k | — | — | 7/10 | 11mo ago |
|
WangRongsheng/awesome-LLM-resources
|
8.7k | — | — | 7/10 | 10h ago |
|
xlite-dev/Awesome-LLM-Inference
|
5.4k | — | Python | 7/10 | 2w ago |
 patchy631/ai-engineering-hub
|
36.4k | — | Jupyter Notebook | 7/10 | 1mo ago |
Broader scope covering all LLM topics; 26,969 stars versus 6,896. Awesome-LLM-Strawberry wins on depth and focus for reasoning specifically, but loses on breadth and likely ongoing maintenance energy.
More narrowly focused on RL for large reasoning models; 2,467 stars. Appears more academically oriented and possibly more current on the RL-for-reasoning literature thread.
General LLM resources aggregator with 8,585 stars. Less focused on reasoning specifically; serves a wider but less specialized audience.
Notebook-driven, practitioner-focused with 35,988 stars. Serves a different need — hands-on code rather than paper curation — so overlap is limited.
Focuses on inference efficiency rather than reasoning techniques; 5,355 stars. Partially overlapping audience but different problem framing.
