Tag
Research tools
5 reviewed tools: 4 free or open source, 1 paid or paid-led.
Order: free and open source first, then hosted products with a real free tier, then paid-led; catalog order inside each group, with quiet or archived repos last. Not a score, and no tool pays for its place.
5 tools
- SWE-agentMIT research OSS coding agent; BYOK; mini is successor
Free / OSS · OSS/MIT · 20.6k GitHub stars
VerdictMIT research SWE harness. Prefer mini-swe-agent going forward per maintainers.
CatchVs OpenHands (productized self-host SWE) and Devin (hosted autonomous): SWE-agent is research OSS, not a polished product seat. Maintainers recommend mini-swe-agent as the successor (simpler control flow; active development). For SWE-bench numbers, read upstream.
- CAMELApache-2.0 multi-agent framework — scaling laws of agents
Free / OSS · Apache-2.0 · 17.9k GitHub stars
VerdictApache-2.0 multi-agent framework. Scale agent societies; Eigent is a different product.
CatchVs CrewAI / AG2 (product multi-agent orchestration) and OpenHands / SWE-agent (coding agents): CAMEL is multi-agent research/framework breadth, not a turnkey SWE seat. Eigent is a separate commercial cousin — do not treat CAMEL OSS as Eigent cloud. You own model spend and orchestration choices.
- OpenResearchMIT coding agents → research agents (Rust)
Free / OSS · OSS/MIT · 7.3k GitHub stars
VerdictMIT research-agent workspace. Coding agents → autoresearch; not Spec Kit/DSPy/Promptfoo.
CatchVs Spec Kit (spec-driven coding workflow), DSPy (prompt/program optimizers), Promptfoo (eval/red-team): OpenResearch is a research-agent harness and experiment ledger around existing coding agents — not a spec kit, not a prompt compiler, not an eval suite. You still pay (or self-host) the coding agent and any remote compute you dispatch.
- Open-Science (AIPOCH)Apache-2.0 agents research workbench — brew cask / download free
Free / OSS · Apache-2.0 · 5.2k GitHub stars
VerdictApache agents research workbench when you want AIPOCH’s free local lane — not OpenResearch’s Rust research agents, not WeKnora RAG.
CatchVs OpenResearch / WeKnora: OpenResearch is MIT coding-agents→research-agents (Rust); WeKnora is Tencent’s MIT self-host RAG/knowledge platform — Open-Science’s catch is the AIPOCH agents research workbench surface, not Rust research-agent rewrite and not RAG-as-platform. Star growth (~21 a week) is healthy, but the product UX is still early — pin a release and verify the brew/cask path before fleet install.
- ScryHosted freemium MCP + SQL over public-record corpora
Paid-led · Freemium
VerdictFreemium public-record research API. MCP+SQL over an indexed corpus; not Firecrawl, Crawl4AI, or an OSS database.
CatchResearcher is non-commercial; Team starts from $2k. This is a hosted data product, not an OSS dump you can fork and self-host. Vs Firecrawl / Crawl4AI / Context7: Scry is indexed corpus + SQL, not page fetch, crawl, or documentation packing. Check corpus coverage, freshness, and commercial terms before treating a query as evidence.
FAQ
- What is the top free pick among Research tools?
- SWE-agent leads the Stack Gems list for Research tools. Verdict: MIT research SWE harness. Prefer mini-swe-agent going forward per maintainers.
- How many Research tools does Stack Gems review?
- 5: 4 free or open source, 1 paid or paid-led. Each review has a verdict and an honest Catch; no tool pays for its spot.
- Which Research tools can I self-host for free?
- SWE-agent, CAMEL, OpenResearch and Open-Science (AIPOCH).
- Which Research tools have the most GitHub stars?
- SWE-agent (20.6k), CAMEL (17.9k), OpenResearch (7.3k), per GitHub on 2026-10-19. Stars measure attention, not fit; read the Catch.
- What is the catch with SWE-agent?
- Vs OpenHands (productized self-host SWE) and Devin (hosted autonomous): SWE-agent is research OSS, not a polished product seat.