DiracvsQwen Code
Generated from the two spec rows. Green marks the better value where a spec has a clear direction. Everything else is just different.
Dirac
Dirac · Terminal agent
#140OSS
Panel
6.81 spec wins
- Reliability
- 6.2
- Usefulness
- 7.2
- Cost
- 7.7
- Longevity
- 6.0
“Its browser control works by screenshots and coordinates, which is how we automated the web before anyone thought to ask the page.”
Qwen Code
Alibaba Cloud (Qwen) · Terminal agent
#19OSSMCP
Panel
6.84 spec wins
- Reliability
- 6.5
- Usefulness
- 6.8
- Cost
- 7.3
- Longevity
- 6.3
“Ships a Java SDK, which tells you precisely which enterprises Alibaba expects to pay for it.”
Spec by spec
| Spec | Dirac | Qwen Code |
|---|---|---|
| Architecture | ||
| Category | Terminal agent | Terminal agent |
| Runs | local | local |
| Platforms | macos, linux, windows | macos, linux, windows |
| Context window | not documented | not documented |
| Protocols | ||
| MCP client | No | Yes |
| MCP server | No | No |
| Capabilities | ||
| Runs terminal commands | Yes | Yes |
| Multi-file edits | Yes | Yes |
| Git operations | YesWork can be isolated in a git worktree with integration and cleanup controls. | Yes |
| Browser control | YesBrowser interaction is screenshot and coordinate based. | YesThe first-party @qwen-code/chrome-bridge extension tunnels cdp_* frames from the `qwen serve` daemon into chrome.debugger so the agent drives a real Chrome tab, but console/network/screenshot/click tools need an external CDP MCP adapter set via QWEN_CDP_MCP_COMMAND, since none is bundled. |
| Sandboxed execution | No | YesTwo real isolation modes - macOS Seatbelt via sandbox-exec (default `permissive-open` profile) and container-based Docker/Podman with a published sandbox image (https://qwenlm.github.io/qwen-code-docs/en/users/features/sandbox/). |
| Multi-agent orchestration | Yes | Yes |
| Headless / CI mode | No | Yes |
| Models | ||
| Backbone | Anthropic, OpenAI, OpenRouter, Gemini, Groq, Mistral, any OpenAI-compatible endpoint | Qwen3-Coder, OpenAI-compatible, Claude, Gemini |
| Bring your own model | Yes | Yes |
| Local models | Yes | Yes |
| Cost | ||
| Pricing model | byok | byok |
| Starts at | $0/mo | n/a |
| Free tier | Yes | No |
| Bring your own key | Yes | Yes |
| Openness | ||
| Open source | Yes | Yes |
| License | Apache-2.0 | Apache-2.0 |
| GitHub stars | 1,518 | 28,271 |
Which one would each critic pick
| Critic | Dirac | Qwen Code | Pick |
|---|---|---|---|
| El Juez | — | — | not enough reviews |
| El Amigo | 7.0 | 7.0 | no preference |
| El Crítico | 6.3 | 6.3 | no preference |
| El Profesor | 7.3 | 6.0 | Dirac — Edits are anchored by content hash rather than by line number, so a stale edit fails loudly instead of applying itself to the wrong region of a changed file. |
| La Inversora | 6.5 | 6.8 | Qwen Code — Alibaba killed the Qwen OAuth free tier on 2026-04-15, the classic subsidy-to-subscription turn, and 27,000 stars now have to pay someone. |
| La Jefa | 5.8 | 5.8 | no preference |
| El Hacker | 7.8 | 8.8 | Qwen Code — Apache-2.0, OPENAI_BASE_URL pointed at my own box, modelProviders in settings.json, and MCP servers with a command or an httpUrl; the whole thing is mine to fork. |
Picks are derived from each critic's own scores. Humans vote on matchups on the duels page.