CodewhalevsFuXi
Generated from the two spec rows. Green marks the better value where a spec has a clear direction. Everything else is just different.
Codewhale
Hmbown · Terminal agent
#181OSSMCP
Panel
5.83 spec wins
- Reliability
- 5.5
- Usefulness
- 5.5
- Cost
- 7.3
- Longevity
- 4.7
“Installs on Android through Termux, so you can run a coding agent on the phone you use to ignore your coding agent.”
FuXi
FUXI · Terminal agent
#90MCP
Panel
5.82 spec wins
- Reliability
- 5.0
- Usefulness
- 6.5
- Cost
- 7.3
- Longevity
- 4.3
“More than fifty built-in tools in one binary, which is either thoroughness or the longest permission prompt in this category.”
Spec by spec
| Spec | Codewhale | FuXi |
|---|---|---|
| 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 | Yes | Yes |
| MCP server | No | Yes |
| Capabilities | ||
| Runs terminal commands | Yes | Yes |
| Multi-file edits | Yes | Yes |
| Git operations | Yes | Yes/commit and /review create commits, review code and open pull requests. |
| Browser control | No | Yes |
| Sandboxed execution | Yes | No |
| Multi-agent orchestration | Yes | Yes |
| Headless / CI mode | Yes | Yes`fuxi setup-token` is documented specifically for headless and CI use. |
| Models | ||
| Backbone | hosted providers, Ollama, vLLM, SGLang | any OpenAI-compatible provider, Anthropic, Gemini, Amazon Bedrock, Google Vertex |
| Bring your own model | Yes | Yes |
| Local models | Yes | YesA custom base_url is documented through config, the FUXI_BASE_URL environment variable and the -b/--base-url flag, which covers a locally hosted OpenAI-compatible server. |
| Cost | ||
| Pricing model | byok | byok |
| Starts at | $0/mo | $0/mo |
| Free tier | Yes | Yes |
| Bring your own key | Yes | Yes |
| Openness | ||
| Open source | Yes | No |
| License | MIT | proprietary |
| GitHub stars | 41,039 | 3,350 |
Which one would each critic pick
| Critic | Codewhale | FuXi | Pick |
|---|---|---|---|
| El Juez | — | — | not enough reviews |
| El Amigo | 6.5 | 6.5 | no preference |
| El Crítico | 5.5 | 5.3 | Codewhale — It was deepseek-tui until recently and carries the old config and sessions forward; the rename is fresh, and the project is one maintainer wide. |
| El Profesor | 5.8 | 6.3 | FuXi — Classifying shell commands by parsing them into a syntax tree is the correct method; pattern matching on command strings is what everyone else does and it is defeated by quoting. |
| La Inversora | 3.8 | 5.3 | FuXi — A free binary with an optional managed model service is a funnel, and three thousand stars on a repository that ships no code is a marketing result. |
| La Jefa | 5.5 | 5.0 | Codewhale — A single-maintainer terminal agent with shell completions and a non-interactive mode, no SSO, no admin and no vendor; approved with conditions for engineers who already hold keys. |
| El Hacker | 7.5 | 6.5 | Codewhale — MIT Rust, Ollama, vLLM and SGLang for local models, a config.example.toml to copy, MCP servers attachable, and builds from Cargo, Nix, Docker or Scoop; it runs offline. |
Picks are derived from each critic's own scores. Humans vote on matchups on the duels page.