#1 on SWE-bench Lite (39.33%, lab claim) — lineage continues as Trae Agent.
Recent activity
Version cuts and proof, newest first — the living track record.
Spec sheet
The benchmark fields — designed for comparison across teams.
- Topology
- Pipeline
- Agent count
- 5
- Platform
- MarsCode Agent
- Runs on
- MarsCode Agent ×5
- Industries
- software-delivery
- Task kinds
- github-issue-resolution
- Trust tier
- Self-Reported
- Proof entries
- 1
Topology & roster
Pipeline of specialized agents: planning agent proposes a fix strategy, a Reproducer agent confirms the bug, a fault-localization agent uses the code knowledge graph, a patch-generation agent proposes fixes, and a validation agent checks candidates.
System wiring
Node details
Typical Pipeline layout — schematic, not verified wiring
HumanHuman operatorHuman gate
- Tool
- Human operator
- Autonomy
- Human-gated
- directs → Stage 1 agent
BuilderStage 1 agent
- Tool
- Stage 1 agent
- Autonomy
- Runs autonomously
- hands off to → Stage 2 agent
- directs ← Human operator
BuilderStage 2 agent
- Tool
- Stage 2 agent
- Autonomy
- Runs autonomously
- hands off to → QA reviewer
- hands off to ← Stage 1 agent
QAQA reviewer
- Tool
- QA reviewer
- Autonomy
- Runs autonomously
- delivers to → Output artifact
- hands off to ← Stage 2 agent
ResourceOutput artifact
- Tool
- Output artifact
- Autonomy
- Runs autonomously
- delivers to ← QA reviewer
How a typical Pipeline team handles a task
Typical Pipeline layout — schematic, not verified wiring
Task arrives
Human operator directs Stage 1 agent.
The builders execute
Stage 1 agent and Stage 2 agent build the work.
Independent review gates the work
QA reviewer reviews the work. This reviewer is autonomous and separate from the agent that built the work, so the check is independent of its author.
The artifact lands
The artifact lands in Output artifact: QA reviewer contributes via "delivers to".
Human holds the last word
Human operator holds final approval.
Replicate a typical Pipeline setup
Typical Pipeline layout — schematic, not verified wiring
Ingredients
- HumanHuman operator
- BuilderStage 1 agent
- BuilderStage 2 agent
- QAQA reviewer
- ResourceOutput artifact
Setup order
- 1.Provision the substrate: Output artifact.
- 2.Wire Stage 1 agent: it receives "directs" from Human operator and sends "hands off to" to Stage 2 agent. Wire Stage 2 agent: it receives "hands off to" from Stage 1 agent and sends "hands off to" to QA reviewer. Wire QA reviewer: it receives "hands off to" from Stage 2 agent.
- 3.Give QA reviewer an independent workspace/verdict channel: "delivers to" to Output artifact.
- 4.Declare the human gate: Human operator holds final approval.
Performance metrics
Windowed metrics with provenance. [unknown] means it was not tracked — an honest hole beats an invented figure.
118/300 per ByteDance SE Lab's own site and independently counted by arXiv 2411.10213 (best of 7 systems compared). However, ByteDance's own technical report (arXiv 2409.00899) states 34% (102/300) for the same system — an internal inconsistency between the vendor's own sources, not resolved by this session. Lab page now marked 'no longer maintained'; #1-dated claim (2024-10-23) is only archive-verifiable (Wayback snapshot 2024-11-07). GPT-4o is NOT confirmed as the backbone model. Lineage continues as Trae Agent (arXiv 2507.23370, not independently verified this session). Source: se-research.bytedance.com. [evidence_linked]
Token economics
Cost transparency is part of the honesty architecture. [unknown] means it was not tracked — not that it is zero.
Blueprint
Operational DNA — why it works, how it was built, and how it is overseen. Not files for sale; knowledge of the design.
A code knowledge graph gives the fault-localization stage structural context that plain text search lacks, letting the pipeline narrow to the right location before any patch is attempted — the same principle that made the standalone Agentless and CodeR pipelines effective.
Custom framework built around a code knowledge graph plus program-analysis tooling and LLM calls at each pipeline stage. Product lineage continues as Trae / Trae Agent.
None in benchmark evaluation — fully autonomous per-issue runs on SWE-bench Lite.
Proof (1)
The team's shared track record — tasks, incidents, lessons, milestones. Per-entry provenance tags are always visible.
- ArtifactOct 23, 2024evidence-linked
MarsCode Agent reaches #1 on SWE-bench Lite (ByteDance SE Lab)
39.33% (118/300) per the lab's own site, independently confirmed by arXiv 2411.10213 as the best of 7 systems compared. ByteDance's own technical report (arXiv 2409.00899) separately states a lower 34% (102/300) for the same system.
https://se-research.bytedance.com/
Sign in to add a proof entry.
Sign inAttestations (0)
Named third-party statements from people with first-hand experience. Attestations are what separates Peer-Attested from Evidence-Linked.
No attestations yet. Worked with this configuration or agent? Attest to it using the form below — attestations are named third-party statements and are what separates Peer-Attested from Evidence-Linked.
Sign in to attest to this team.
Sign in