BitterbotvsFractal
Generated from the two spec rows. Green marks the better value where a spec has a clear direction. Everything else is just different.
Bitterbot
Bitterbot-AI · Terminal agent
#264OSS
Panel
4.32 spec wins
- Reliability
- 2.8
- Usefulness
- 4.2
- Cost
- 6.0
- Longevity
- 4.2
“It runs code on your local machine without a sandbox, which is one way to give an AI persistent memory.”
Fractal
Trampoline AI · Terminal agent
#224OSS
Panel
4.63 spec wins
- Reliability
- 3.7
- Usefulness
- 4.2
- Cost
- 6.0
- Longevity
- 4.7
“The runtime is described as self-harnessed, which is a very confident way to say the loop calls itself.”
Spec by spec
| Spec | Bitterbot | Fractal |
|---|---|---|
| Architecture | ||
| Category | Terminal agent | Terminal agent |
| Runs | local | local |
| Platforms | macos, linux, windows | macos |
| Context window | not documented | not documented |
| Protocols | ||
| MCP client | No | No |
| MCP server | No | No |
| Capabilities | ||
| Runs terminal commands | Yes | Yes |
| Multi-file edits | No | Yes |
| Git operations | No | No |
| Browser control | Yes | No |
| Sandboxed execution | No | Yes |
| Multi-agent orchestration | No | Yes |
| Headless / CI mode | No | No |
| Models | ||
| Backbone | predict-rlm | |
| Bring your own model | Yes | Yes |
| Local models | No | No |
| Cost | ||
| Pricing model | free | byok |
| Starts at | n/a | $0/mo |
| Free tier | Yes | Yes |
| Bring your own key | Yes | Yes |
| Openness | ||
| Open source | Yes | Yes |
| License | MIT | MIT |
| GitHub stars | 2,463 | 200 |
Which one would each critic pick
| Critic | Bitterbot | Fractal | Pick |
|---|---|---|---|
| El Juez | — | — | not enough reviews |
| El Amigo | 4.5 | 4.5 | no preference |
| El Crítico | 5.3 | 4.3 | Bitterbot — Bitterbot runs code without a sandbox, creating a direct path for a compromised model to access your local file system. |
| El Profesor | 2.0 | 5.3 | Fractal — The agent is structured as a recursive language model rather than a flat loop, so decomposition is the control flow instead of a prompt convention. |
| La Inversora | 5.0 | 4.8 | Bitterbot — Bitterbot's 'P2P skills economy' is a novel idea, but a business model based on user-to-user transactions without a clear company take is a classic venture-funded experiment. |
| La Jefa | 2.5 | 3.5 | Fractal — The authors call it early alpha, which ends the conversation before the questions about single sign-on, audit logging and unattended runs even start. |
| El Hacker | 6.5 | 5.5 | Bitterbot — Bitterbot is a local-first agent with a solid MIT license and BYOK, but the lack of a code sandbox is a significant risk for a tool that can execute commands. |
Picks are derived from each critic's own scores. Humans vote on matchups on the duels page.