World-3 · web and tooling · World-3 collaboration. kimi-k2.7-code on Kimi CLI, 26 agents, connected: 1,252 sessions, 253 agent hours, 5.49 B tokens, $1,349 at list prices. From data/per-release-v3/exp8 in the paper’s figures repository at 9014a8cc, fetched from co-evolving/experiment-8 at 81f4ed75 on Hugging Face. The states and releases come from the checkpoint co-evolving/experiment-8 @ f12d7fd8 (branch main). The score is rebuilt from the forge’s history, one state every two hours, each task carried at its last measured value.
Scarry 1.124 +12.4% over the seed at the last reading, after 1,240 sessions −0.1 pp from the previous reading
Sstrict 1.049 a package with a failed check or benchmark scores 1.0 2 packages failed at the last reading
Movers +7 / −2 tasks beyond their noise floor at the last reading
Readings 75 latest Aug 13, 17:06 UTC
Run 1,252 sessions completed work sessions eval set v3.x · speedup over the seed sessions time
+5.0% +10.0% seed 0 200 400 600 800 1000 1200 completed sessions Release: aiosignal v1.4.1 Release: fastapi 0.141.2 Release: jsonschema-specifications v2025.9.2 Release: attrs 26.2.0 Release: fastapi 0.141.3 Release: aiosignal v1.4.2 Release: pydantic-settings v2.14.3 Release: fastapi 0.141.4 Release: packaging 26.4 Releases: pydantic v2.13.5, pydantic v2.46.5 Release: fastapi 0.141.5 Release: rich v15.0.1 Release: pydantic-settings v2.14.4 Release: h11 0.17.0 Release: fastapi 0.141.6 Release: packaging 26.5 Releases: referencing v0.37.1, cloudpickle v3.1.3 Release: attrs 26.3.0 Release: jsonschema v4.26.1 Release: pydantic-settings v2.14.5 Release: fastapi 0.141.7 Release: rich v15.0.2 Release: packaging 26.6 Release: referencing v0.37.2 Release: cloudpickle v3.1.4 Release: jsonschema-specifications v2025.9.3 Release: attrs 26.4.0 Release: pydantic-settings v2.14.6 Release: h11 0.18.0 Release: fastapi 0.141.8 Release: rich v15.0.3 Release: packaging 26.7 Release: referencing v0.37.3 Release: jsonschema-specifications v2025.9.4 Release: jsonschema v4.26.2 Release: attrs 26.5.0 Release: pydantic-settings v2.14.7 Release: h11 v0.19.0 Release: fastapi 0.141.9 Release: rich v15.0.4 Release: referencing v0.37.4 Release: attrs 26.6.0 Release: pydantic-settings v2.14.8 Release: fastapi 0.141.10 Release: rich v15.0.5 Release: referencing v0.37.5 Release: pyyaml 6.0.4 Release: pydantic-settings v2.14.9 Release: packaging 26.8 Release: referencing v0.37.6 Release: attrs 26.7.0 Release: rich v15.0.6 Release: h11 v0.20.0 Release: referencing v0.37.7 Release: pydantic-settings v2.14.10 Release: attrs 26.8.0 Release: rich v15.0.7 Release: fastapi 0.141.11 Releases: pydantic v2.13.6, pydantic v2.46.6 Release: jsonschema v4.26.3 Release: referencing v0.37.8 Release: pyyaml 6.0.5 Release: pydantic-settings v2.14.11 Release: attrs 26.9.0 Release: h11 v0.21.0 Release: fastapi 0.141.12 Release: packaging 26.9 Releases: referencing v0.37.9, pyyaml 6.0.6 Release: cloudpickle v3.1.5 Releases: aiosignal v1.4.3, attrs 26.10.0 Releases: rich v15.0.8, h11 v0.22.0 Release: fastapi 0.141.13 Releases: pydantic v2.13.7, pydantic v2.46.7, packaging 26.10 Release: jsonschema v4.26.4 Release: pyyaml 6.0.7 Release: attrs 26.11.0 Release: aiohttp v3.14.4 Releases: rich v15.0.9, h11 v0.23.0 Release: packaging 26.11 Release: jsonschema v4.26.5 Release: pyyaml 6.0.8 Release: aiosignal v1.4.4 Release: h11 v0.24.0 Release: rich v15.0.10 Release: referencing v0.37.10 Releases: fastapi 0.141.14, packaging 26.12 Release: jsonschema v4.26.6 Release: cloudpickle v3.1.6 Release: pydantic-settings v2.14.12 Release: attrs 26.12.0 Release: aiohttp v3.14.5 Release: rich v15.0.11 Release: h11 v0.25.0 Release: packaging 26.13 Releases: pydantic-settings v2.14.13, attrs 26.13.0 Releases: referencing v0.37.11, rich v15.0.12 Release: h11 v0.26.0 Release: jsonschema v4.26.7 Release: pydantic-settings v2.14.14 Release: attrs 26.14.0 Release: referencing v0.37.12 Releases: rich v15.0.13, fastapi 0.141.15 Release: h11 v0.27.0 Release: packaging 26.14 Release: cloudpickle v3.1.7 Release: pydantic-settings v2.14.15 Release: attrs 26.15.0 Release: referencing v0.37.13 Releases: rich v15.0.14, fastapi 0.141.16 Release: jsonschema v4.26.8 Release: cloudpickle v3.1.8 Release: attrs 26.16.0 Release: rich v15.0.15 Releases: referencing v0.37.14, fastapi 0.141.17 Release: packaging 26.15 Release: jsonschema v4.26.9 Release: pyyaml 6.0.9 Release: pydantic-settings v2.14.16 Release: cloudpickle v3.1.9 Release: attrs 26.17.0 Releases: rich v15.0.16, referencing v0.37.15 Release: fastapi 0.141.18 Release: packaging 26.16 Release: pyyaml 6.0.10 Release: attrs 26.18.0 Releases: referencing v0.37.16, rich v15.0.17 Release: h11 v0.28.0 Release: pydantic-settings v2.14.17 Release: attrs 26.19.0 Release: rich v15.0.18 Release: fastapi 0.141.19 Releases: h11 v0.29.0, jsonschema v4.26.10 Release: packaging 26.17 Release: attrs 26.20.0 Release: rich v15.0.19 Release: fastapi 0.141.20 Release: h11 v0.30.0 Release: jsonschema v4.26.11 Release: pydantic-settings v2.14.18 Release: attrs 26.21.0 Release: rich v15.0.20 Releases: fastapi 0.141.21, h11 v0.31.0 Release: jsonschema v4.26.12 Release: pydantic v2.13.8 Releases: cloudpickle v3.1.10, attrs 26.22.0 Release: referencing v0.37.17 Release: rich v15.0.21 Release: fastapi 0.141.22 Releases: packaging 26.18, jsonschema v4.26.13 Releases: pyyaml 6.0.11, pydantic-settings v2.14.19 Release: aiohttp v3.14.6 Release: referencing v0.37.18 Release: rich v15.0.22 Release: fastapi 0.141.23 Release: jsonschema v4.26.14 Release: pydantic-settings v2.14.20 Release: attrs 26.23.0 Release: referencing v0.37.19 Release: rich v15.0.23 Release: fastapi 0.141.24 Releases: packaging 26.19, jsonschema v4.26.15 Release: pyyaml 6.0.12 Release: aiohttp v3.14.7 Release: cloudpickle v3.1.11 Release: h11 v0.32.0 Release: rich v15.0.24 Release: pydantic-settings v2.14.21 Releases: pyyaml 6.0.13, jsonschema v4.26.16 Release: aiohttp v3.14.8 Release: referencing v0.37.20 Releases: fastapi 0.141.25, packaging 26.20 Release: pydantic-settings v2.14.22 Releases: pyyaml 6.0.14, jsonschema v4.26.17 Release: attrs 26.24.0 Release: referencing v0.37.21 Release: rich v15.0.25 Release: packaging 26.21 Release: pyyaml 6.0.15 Release: jsonschema v4.26.18 Releases: pydantic v2.13.9, pydantic v2.46.9 Release: h11 v0.33.0 Release: referencing v0.37.22 Release: rich v15.0.26 Release: pydantic-settings v2.14.23 Release: pyyaml 6.0.16 Release: jsonschema v4.26.19 Release: referencing v0.37.23 Releases: rich v15.0.27, packaging 26.22 Release: pyyaml 6.0.17 Release: aiohttp v3.14.9 Releases: pydantic v2.13.10, pydantic v2.46.10 Releases: h11 v0.34.0, cloudpickle v3.1.12 Release: referencing v0.37.24 Release: pydantic-settings v2.14.24 Release: jsonschema v4.26.20 Release: attrs 26.25.0 Release: packaging 26.23 Release: pydantic-settings v2.14.25 Release: aiohttp v3.14.10 Releases: pydantic v2.13.11, pydantic v2.46.11 Releases: pyyaml 6.0.18, rich v15.0.28, referencing v0.37.25 Release: pydantic-settings v2.14.26 Releases: jsonschema v4.26.21, aiohttp v3.14.11 Release: packaging 26.24 Release: rich v15.0.29 Release: referencing v0.37.26 Release: pydantic-settings v2.14.27 Release: pydantic v2.13.12 Release: h11 v0.35.0 Release: attrs 26.26.0 Release: rich v15.0.30 Release: referencing v0.37.27 Release: pydantic-settings v2.14.28 Release: attrs 26.27.0 Release: packaging 26.25 Release: rich v15.0.31 Release: fastapi 0.141.26 Release: jsonschema v4.26.22 Release: pydantic v2.13.13 Release: attrs 26.28.0 Release: packaging 26.26 Release: referencing v0.37.28 Release: rich v15.0.32 Releases: pydantic-settings v2.14.29, fastapi 0.141.27 Release: jsonschema v4.26.23 Release: h11 v0.36.0 Releases: attrs 26.29.0, packaging 26.27 Release: pydantic-settings v2.14.30 Release: rich v15.0.33 Release: jsonschema v4.26.24 Release: pydantic v2.13.14 Release: h11 v0.37.0 Release: attrs 26.30.0 Release: pyyaml 6.0.19 Release: pydantic-settings v2.14.31 Releases: fastapi 0.141.28, propcache v0.5.7 Release: uvicorn 0.52.4 Release: jsonschema v4.26.25 Releases: attrs 26.31.0, rich v15.0.34 Release: pydantic-settings v2.14.32 Releases: referencing v0.37.29, aiohttp v3.14.12 Release: fastapi 0.141.29 Release: jsonschema v4.26.26 Releases: attrs 26.32.0, pyyaml 6.0.20 Release: packaging 26.28 Release: rich v15.0.35 Release: pydantic-settings v2.14.33 Release: referencing v0.37.30 Releases: fastapi 0.141.30, jsonschema v4.26.27 Releases: rich v15.0.36, packaging 26.29, cloudpickle v3.1.13 Release: pydantic-settings v2.14.34 Release: jsonschema v4.26.28 Releases: h11 0.38.0, pytest 9.1.8 Release: attrs 26.33.0 Release: rich v15.0.37 Release: cloudpickle v3.1.14 Release: aiohttp v3.14.13 Releases: pydantic-settings v2.14.35, iniconfig v2.3.13 Release: jsonschema v4.26.29 Release: attrs 26.34.0 Release: packaging 26.30 Release: rich v15.0.38 Release: h11 0.39.0 Release: frozenlist v1.9.14 Release: packaging 26.31 Release: rich v15.0.39 Release: pydantic-settings v2.14.36 Release: jsonschema v4.26.30 Release: attrs 26.35.0 Release: rich v15.0.40 Release: pydantic-settings v2.14.37 Release: referencing v0.37.31 Release: pluggy 1.35.0 Release: jsonschema v4.26.31 Release: attrs 26.36.0 Release: packaging 26.32 Release: aiohttp v3.14.14 Release: cloudpickle v3.1.15 Releases: referencing v0.37.32, fastapi 0.141.31 Release: pygments 2.20.20 Release: attrs 26.37.0 Releases: yarl v1.24.43, pydantic-settings v2.14.38 Release: aiohttp v3.14.15 Release: tibs v0.5.42 Release: multidict v6.7.22 Releases: rich v15.0.41, attrs 26.38.0 Scarry at each readingSstrict , where it differs a package failed a check or a benchmark 334 releases measuredBreakdown after 1,240 sessions · Aug 13, 17:06 UTC Scarry 1.124 · Sstrict 1.049 · +6.5% over the tasks with a value · 15/20 tasks measured
kedro · io-bound-runners +36.1% kedro · data-catalog +35.8% kedro · omega-config-loader +20.8% kedro · session-lifecycle +19.4% openapi-spec-validator · validate-synthetic-specs +16.9% openapi-spec-validator · validate-real-world-specs +13.6% dascore · io-benchmarks +0.6% navigator-api · singleton-construction +0.2% tqdm · alternative-progress-libraries +0.1% dascore · patch-benchmarks 0.0% dascore · spool-benchmarks 0.0% openapi-schema-validator · compiled-validation-throughput benchmark failed openapi-schema-validator · first-validation benchmark failed openapi-schema-validator · helper-validation-checked-throughput benchmark failed openapi-schema-validator · helper-validation-trusted-throughput benchmark failed prefect · cold-imports checks failed tqdm · tqdm-loop-overhead 0.0% mitmproxy · proxy-layer-roundtrips −0.1% navigator-api · base-app-handler-construction −0.7% openapi-schema-validator · schema-compilation −27.3%