github.com/evotai/evot @v2026.7.6
Chat with this repoWhat it actually does AI analysis from the code graph — generated when you open this
README
Evot
An agent engine that completes complex, long-running work with minimal tokens and maximum quality.
Every gain measured under a rigorous trace + eval framework — earned through relentless iteration, never guessed at.
News · Benchmark · Why · Dashboard · Install · Quickstart · Dev
📢 News
- 2026-07-03 [REPL]
/copy— copy the last agent message's Markdown source to the clipboard. - 2026-06-30 [Distill]
evot distill— generate verified SFT/RL datasets by driving evot as a teacher. - 2026-06-16 [REPL] Shift+Tab cycles reasoning effort; persisted per session.
- 2026-06-05 [Dashboard] Built-in web dashboard — server metrics, sessions, usage, and tool traces.
- 2026-05-30 [Engine] Major refactor — four-pass compaction, pi-aligned parallel tools, leaner core.
Benchmark
Same task, same eval environment, different models. evot completes the work with fewer tokens, less time, and lower cost — on both frontier and open-source models.
| Claude Opus 4.6 | DeepSeek V4 Pro |
 |
 |
Task: Fix a real bug in serde_json (issue #979) — investigate root cause, apply fix, write regression test, verify all tests pass.
| Model | Metric | evot | claude-code | Difference |
|---|---|---|---|---|
| Opus 4.6 | Cost | $2.24 | $6.16 | 64% cheaper |
| Opus 4.6 | Time | 2m 56s | 3m 51s | 24% faster |
| Opus 4.6 | Input tokens | 574.8K | 1.5M | 62% fewer |
| DeepSeek V4 Pro | Cost | $0.02 | $0.07 | 67% cheaper |
| DeepSeek V4 Pro | Time | 6m 10s | 16m 34s | 63% faster |
| DeepSeek V4 Pro | Input tokens | 42.9K | 133.8K | 68% fewer |
All agents produce correct, passing code. The difference is how they manage context.
Why is evot faster and cheaper?
Give the LLM less context, but higher-quality context. Where other agents call the LLM to summarize when context overflows — burning extra tokens and time — evot uses zero LLM calls for context management:
- Algorithmic compaction — a four-pass Rust pipeline (Reclaim → Shrink → Collapse → Evict) runs in microseconds between turns. Images downgrade to path references; old turns collapse to one-line summaries.
- Spill to disk — large tool results write to disk with a short preview. The model re-reads on demand instead of carrying megabytes in context.
- Compaction markers — structured metadata (files modified, conclusions, environment state) survives compaction, so progress is never lost.
Every gain is earned under a rigorous trace + eval framework, not guessed at. Each engine change is measured against live traces and a reproducible benchmark pipeline — the same real-world tasks run against Claude Code and Codex (latest versions) — before it ships. Token usage, cost, time, and success rate must improve or hold. Relentless trial and iteration, where the numbers decide what stays. Continuous improvement, no regression.
Dashboard
Evot ships with a built-in web dashboard for real-time observability: server resource usage, all connected sessions, and per-session detail — token usage, tool call sequences, and span-level traces.
| Overview — server metrics & sessions | Session detail — usage & tool traces |
 |
 |
Installation
One-liner (recommended)
curl -fsSL https://evot.ai/install | sh
From source
git clone https://github.com/evotai/evot.git
cd evot
make setup && make install
evot
Quickstart
1. Set your API key
Create ~/.evotai/evot.env:
# Anthropic (default)
EVOT_LLM_ANTHROPIC_API_KEY=sk-ant-...
EVOT_LLM_ANTHROPIC_BASE_URL=your-anthropic-base-url
EVOT_LLM_ANTHROPIC_MODEL=claude-opus-4-6
# Multiple models: EVOT_LLM_ANTHROPIC_MODEL=claude-sonnet-4-6,claude-opus-4-6
# Or OpenAI
# EVOT_LLM_OPENAI_API_KEY=sk-...
# EVOT_LLM_OPENAI_BASE_URL=your-openai-base-url/v1
# EVOT_LLM_OPENAI_MODEL=gpt-5.5
# Or DeepSeek (Anthropic-compatible)
# EVOT_LLM_DEEPSEEK_API_KEY=sk-...
# EVOT_LLM_DEEPSEEK_BASE_URL=https://api.deepseek.com/anthropic
# EVOT_LLM_DEEPSEEK_PROTOCOL=anthropic
# EVOT_LLM_DEEPSEEK_MODEL=deepseek-v4-pro
# Or Xiaomi MiMo-V2.5-Pro (Anthropic-compatible)
# EVOT_LLM_XIAOMI_API_KEY=tp-...
# EVOT_LLM_XIAOMI_BASE_URL=https://token-plan-cn.xiaomimimo.com/anthropic
# EVOT_LLM_XIAOMI_PROTOCOL=anthropic
# EVOT_LLM_XIAOMI_MODEL=mimo-v2.5-pro
Use
--model provider:modelfor one-off overrides.
2. Run
evot # interactive REPL
evot -p "summarize today's PRs" # one-shot task
evot -p "review this" -f ./src/main.rs # attach file context
evot -p "continue work" -c # continue latest session in cwd
evot -p "continue work" -r my-session # resume or create session
evot distill --auto --domain "python backend" # generate an SFT/RL dataset
In the REPL:
/helplists commands, Shift+Tab cycles the reasoning effort.
CLI flags & options
| Flag | Description |
|---|---|
-p, --prompt |
Run a single prompt and exit |
-f, --file <path> |
Attach file/directory context (repeatable) |
-c, --continue |
Continue the latest session in the current directory |
-r, --resume <id> |
Resume or create a session |
--model <model> |
Override the configured model |
--env-file <path> |
Path to a custom evot.env |
--skills <dir> |
Add a skills directory (repeatable) |
--verbose |
Enable info-level logging |
Development
make setup # install Rust toolchain, git hooks
make test # all tests (engine + CLI)
make install # compile standalone binary to ~/.evotai/bin/evot
License
Apache-2.0