InternRobotics/InternManip
41.5
Weak · 16 November 2026
27.7k
lines of production code
Python
primary language
1
measurement over time
What this system is
Behavioural changes
1 commit (0 fixes) modifying challenge/scripts
A change to existing behaviour in challenge/scripts — 1 commit, 1 file: “IROS Onsite Competition (\#22)” (refs \#22).
challenge/scripts · medium confidence · unverified
10 commits (1 fix) modifying internmanip/dataset
A change to existing behaviour in internmanip/dataset — 10 commits (1 fix), 10 files: “\[chore\] Optimized the train/eval data for genmanip, ensuring consistency with the IROS competition. \* Merge branch 'develop' into 'public' \* !53 \[feat\] add soft/hard task sr selection \* !52 \[feat\] optimizing parallel eval & update doc \* !51 update the genmanip data downloading \* !50 重新修改slice,支持default to None \* 更新文件: agent\cfg.py \ 更新文件: start\evaluator.py \ 更新文件: base.py \* 更新文件: dp\cfg.py \ 更新文件: dp\_clip.py”; “\[chore\] pass pre-commit hooks”; “\[chore\] pass pre-commit hook”; and 7 more (refs \#22).
internmanip/dataset · medium confidence · unverified
10 commits (5 fixes) modifying internmanip/benchmarks
A change to existing behaviour in internmanip/benchmarks — 10 commits (5 fixs), 108 files: “\[chore\] Optimized the train/eval data for genmanip, ensuring consistency with the IROS competition. \* Merge branch 'develop' into 'public' \* !53 \[feat\] add soft/hard task sr selection \* !52 \[feat\] optimizing parallel eval & update doc \* !51 update the genmanip data downloading \* !50 重新修改slice,支持default to None \* 更新文件: agent\cfg.py \ 更新文件: start\evaluator.py \ 更新文件: base.py \* 更新文件: dp\cfg.py \ 更新文件: dp\_clip.py”; “\[chore\] pass pre-commit hook”; “\[refactor\] Refactor documentation, complete feature implementation, and add challenge setup”; and 7 more.
internmanip/benchmarks · medium confidence · unverified
11 commits (2 fixes) modifying scripts/train
A change to existing behaviour in scripts/train — 11 commits (2 fixs), 1 file: “\[docs\] update readme.md”; “\[chore\] pass pre-commit hooks”; “\[chore\] pass pre-commit hook”; and 8 more.
scripts/train · medium confidence · unverified
15 commits (5 fixes) modifying challenge
A change to existing behaviour in challenge — 15 commits (5 fixs), 3 files: “\[chore\] Optimized the train/eval data for genmanip, ensuring consistency with the IROS competition. \* Merge branch 'develop' into 'public' \* !53 \[feat\] add soft/hard task sr selection \* !52 \[feat\] optimizing parallel eval & update doc \* !51 update the genmanip data downloading \* !50 重新修改slice,支持default to None \* 更新文件: agent\cfg.py \ 更新文件: start\evaluator.py \ 更新文件: base.py \* 更新文件: dp\cfg.py \ 更新文件: dp\clip.py”; “\[chore\] pass pre-commit hooks”; “!48 \[doc\] add quay tutorial in guidelines \ \[fix\] update eval.sh \* \[doc\] update guidelines \* \[doc\] update guidelines”; and 12 more (refs \#8).
challenge · medium confidence · unverified
16 commits (6 fixes) modifying run\_configs/eval
A change to existing behaviour in run\configs/eval — 16 commits (6 fixs), 16 files: “\[chore\] Optimized the train/eval data for genmanip, ensuring consistency with the IROS competition. \ Merge branch 'develop' into 'public' \* !53 \[feat\] add soft/hard task sr selection \* !52 \[feat\] optimizing parallel eval & update doc \* !51 update the genmanip data downloading \* !50 重新修改slice,支持default to None \* 更新文件: agent\cfg.py \ 更新文件: start\evaluator.py \ 更新文件: base.py \* 更新文件: dp\cfg.py \ 更新文件: dp\clip.py”; “\[chore\] pass pre-commit hooks”; “!10 fix gr00t-n1.5 training and genmanip eval bugs \ \[fix\] gr00t-n1.5 training and genmanip eval bugs”; and 13 more.
_run\configs/eval · medium confidence · unverified
16 commits (7 fixes) modifying internmanip/agent
A change to existing behaviour in internmanip/agent — 16 commits (7 fixs), 29 files: “\[chore\] Optimized the train/eval data for genmanip, ensuring consistency with the IROS competition. \* Merge branch 'develop' into 'public' \* !53 \[feat\] add soft/hard task sr selection \* !52 \[feat\] optimizing parallel eval & update doc \* !51 update the genmanip data downloading \* !50 重新修改slice,支持default to None \* 更新文件: agent\cfg.py \ 更新文件: start\evaluator.py \ 更新文件: base.py \* 更新文件: dp\cfg.py \ 更新文件: dp\clip.py”; “fix GenmanipAgent index mapping \ \[fix\] GenmanipAgent index mapping”; “\[chore\] pass pre-commit hooks”; and 13 more (refs \#22).
internmanip/agent · medium confidence · unverified
17 commits (4 fixes) modifying (repo-wide)
A change to existing behaviour in (repo-wide) — 17 commits (4 fixs), 8 files: “\[chore\] Optimized the train/eval data for genmanip, ensuring consistency with the IROS competition. \* Merge branch 'develop' into 'public' \* !53 \[feat\] add soft/hard task sr selection \* !52 \[feat\] optimizing parallel eval & update doc \* !51 update the genmanip data downloading \* !50 重新修改slice,支持default to None \* 更新文件: agent\cfg.py \ 更新文件: start\evaluator.py \ 更新文件: base.py \* 更新文件: dp\cfg.py \ 更新文件: dp\_clip.py”; “\[docs\] update readme.md”; “\[chore\] pass pre-commit hook”; and 11 more (refs \#22).
(repo-wide) · low confidence · unverified
17 commits (6 fixes) modifying internmanip/model
A change to existing behaviour in internmanip/model — 17 commits (6 fixs), 73 files: “\[chore\] Optimized the train/eval data for genmanip, ensuring consistency with the IROS competition. \* Merge branch 'develop' into 'public' \* !53 \[feat\] add soft/hard task sr selection \* !52 \[feat\] optimizing parallel eval & update doc \* !51 update the genmanip data downloading \* !50 重新修改slice,支持default to None \* 更新文件: agent\cfg.py \ 更新文件: start\evaluator.py \ 更新文件: base.py \* 更新文件: dp\cfg.py \ 更新文件: dp\clip.py”; “!69 update open pi0 and gr00t n1.5 \ !68 \[fix\] fix bug ensuring gr00t training \* Merge branch 'develop' into 'public' \* !65 openpi0 \* Merge branch 'develop' into 'public' \* !63 \[fix\] sync gr00t n1\5 eval \ !60 \[fix\] fix typo in key selection and add notes on how augmentation works \* !58 \[Fix\] fix hardcoded gripper key in dataset augmentation \* Merge branch 'develop' into 'public' \* 更新文件: gr00t\_n1\_5\_on\genmanip.py \ !56 merge public to devcelop \* update video encoder and pi0 on genmanip \* Merge branch 'develop' into 'public' \* !53 \[feat\] add soft/hard task sr selection \* !52 \[feat\] optimizing parallel eval & update doc \* !51 update the genmanip data downloading \* !50 重新修改slice,支持default to None \* 更新文件: agent\cfg.py \ 更新文件: start\evaluator.py \ 更新文件: base.py \* 更新文件: dp\cfg.py \ 更新文件: dp\_clip.py”; “\[chore\] pass pre-commit hooks”; and 14 more.
internmanip/model · medium confidence · unverified
2 commits (0 fixes) modifying internmanip/configs, internmanip/trainer
A change to existing behaviour in internmanip/configs, internmanip/trainer — 2 commits, 5 files: “\[chore\] pass pre-commit hook”; “\[dev\] initial commit”.
internmanip/configs, internmanip/trainer · medium confidence · unverified
3 commits (0 fixes) modifying internmanip/demo\_data
A change to existing behaviour in internmanip/demo\_data — 3 commits, 24 files: “\[chore\] pass pre-commit hook”; “\[chore\] cherry-pick commit history”; “\[dev\] initial commit”.
_internmanip/demo\data · medium confidence · unverified
3 commits (1 fix) modifying assets
A change to existing behaviour in assets — 3 commits (1 fix), 2 files: “\[refactor\] Refactor documentation, complete feature implementation, and add challenge setup”; “\[dev\] initial commit”; “Fixed the issue where tune\_llm and tune\_vision had no effect”.
assets · medium confidence · unverified
3 commits (1 fix) modifying internmanip/unit\_test\_script
A change to existing behaviour in internmanip/unit\_test\_script — 3 commits (1 fix), 6 files: “\[dev\] initial commit”; “Delete unnecessary directory”; “fix hard-coded name and paths”.
_internmanip/unit\_test\script · medium confidence · unverified
5 commits (1 fix) modifying internmanip/configs/agent
A change to existing behaviour in internmanip/configs/agent — 5 commits (1 fix), 4 files: “\[chore\] Optimized the train/eval data for genmanip, ensuring consistency with the IROS competition. \* Merge branch 'develop' into 'public' \* !53 \[feat\] add soft/hard task sr selection \* !52 \[feat\] optimizing parallel eval & update doc \* !51 update the genmanip data downloading \* !50 重新修改slice,支持default to None \* 更新文件: agent\cfg.py \ 更新文件: start\evaluator.py \ 更新文件: base.py \* 更新文件: dp\cfg.py \ 更新文件: dp\_clip.py”; “\[chore\] pass pre-commit hook”; “\[dev\] initial commit”; and 2 more (refs \#22).
internmanip/configs/agent · medium confidence · unverified
5 commits (2 fixes) modifying bash\_scripts
A change to existing behaviour in bash\scripts — 5 commits (2 fixs), 3 files: “!10 fix gr00t-n1.5 training and genmanip eval bugs \ \[fix\] gr00t-n1.5 training and genmanip eval bugs”; “\[dev\] initial commit”; “fix hard-coded name and paths”; and 2 more.
_bash\scripts · medium confidence · unverified
5 commits (2 fixes) modifying run\_configs/examples
A change to existing behaviour in run\configs/examples — 5 commits (2 fixs), 1 file: “\[chore\] pass pre-commit hook”; “!11 update demo \ update demo”; “!76 \[fix\] update to v0.1 \* update pi0 \* remove useless comment \* remove hard code \* remove hard code \* remove hard code in internmanip \* !75 update eval demo \* !74 \[fix\] fix genmanip eval bug by update InternUtopia \* !73 \[test\] added gr00t n1\5 test cases for genmanip training and testing \ !52 \[feat\] optimizing parallel eval & update doc”; and 2 more.
_run\configs/examples · medium confidence · unverified
6 commits (1 fix) modifying run\_configs
A change to existing behaviour in run\_configs — 6 commits (1 fix), 1 file: “\[chore\] pass pre-commit hook”; “\[refactor\] Refactor documentation, complete feature implementation, and add challenge setup”; “\[dev\] initial commit”; and 3 more.
_run\configs · medium confidence · unverified
7 commits (0 fixes) modifying internmanip/configs/dataset
A change to existing behaviour in internmanip/configs/dataset — 7 commits, 1 file: “\[chore\] pass pre-commit hook”; “clear and check data-config”; “\[refactor\] Refactor documentation, complete feature implementation, and add challenge setup”; and 4 more (refs \#22).
internmanip/configs/dataset · medium confidence · unverified
7 commits (2 fixes) modifying internmanip/configs/env
A change to existing behaviour in internmanip/configs/env — 7 commits (2 fixs), 4 files: “\[chore\] Optimized the train/eval data for genmanip, ensuring consistency with the IROS competition. \* Merge branch 'develop' into 'public' \* !53 \[feat\] add soft/hard task sr selection \* !52 \[feat\] optimizing parallel eval & update doc \* !51 update the genmanip data downloading \* !50 重新修改slice,支持default to None \* 更新文件: agent\cfg.py \ 更新文件: start\evaluator.py \ 更新文件: base.py \* 更新文件: dp\cfg.py \ 更新文件: dp\_clip.py”; “\[chore\] pass pre-commit hook”; “\[refactor\] Refactor documentation, complete feature implementation, and add challenge setup”; and 4 more (refs \#22).
internmanip/configs/env · medium confidence · unverified
8 commits (3 fixes) modifying challenge/run\_configs/eval
A change to existing behaviour in challenge/run\configs/eval — 8 commits (3 fixs), 2 files: “\[chore\] Optimized the train/eval data for genmanip, ensuring consistency with the IROS competition. \ Merge branch 'develop' into 'public' \* !53 \[feat\] add soft/hard task sr selection \* !52 \[feat\] optimizing parallel eval & update doc \* !51 update the genmanip data downloading \* !50 重新修改slice,支持default to None \* 更新文件: agent\cfg.py \ 更新文件: start\evaluator.py \ 更新文件: base.py \* 更新文件: dp\cfg.py \ 更新文件: dp\clip.py”; “\[refactor\] refactor path”; “!72 \[fix\] fix dp, gr00t, pi0, open pi0 train & eval bugs \ !70 fix dp training and evaluation on the genmanip dataset \* \[fix\] fix bugs for gr00t eval \* !68 \[fix\] fix bug ensuring gr00t training \* Merge branch 'develop' into 'public' \* !65 openpi0 \* Merge branch 'develop' into 'public' \* !63 \[fix\] sync gr00t n1\5 eval \ !60 \[fix\] fix typo in key selection and add notes on how augmentation works \* !58 \[Fix\] fix hardcoded gripper key in dataset augmentation \* Merge branch 'develop' into 'public' \* 更新文件: gr00t\_n1\_5\_on\genmanip.py \ !56 merge public to devcelop \* update video encoder and pi0 on genmanip \* Merge branch 'develop' into 'public' \* !53 \[feat\] add soft/hard task sr selection \* !52 \[feat\] optimizing parallel eval & update doc \* !51 update the genmanip data downloading \* !50 重新修改slice,支持default to None \* 更新文件: agent\cfg.py \ 更新文件: start\evaluator.py \ 更新文件: base.py \* 更新文件: dp\cfg.py \ 更新文件: dp\_clip.py”; and 5 more (refs \#22).
_challenge/run\configs/eval · medium confidence · unverified
8 commits (3 fixes) modifying scripts/eval
A change to existing behaviour in scripts/eval — 8 commits (3 fixs), 5 files: “\[chore\] Optimized the train/eval data for genmanip, ensuring consistency with the IROS competition. \* Merge branch 'develop' into 'public' \* !53 \[feat\] add soft/hard task sr selection \* !52 \[feat\] optimizing parallel eval & update doc \* !51 update the genmanip data downloading \* !50 重新修改slice,支持default to None \* 更新文件: agent\cfg.py \ 更新文件: start\evaluator.py \ 更新文件: base.py \* 更新文件: dp\cfg.py \ 更新文件: dp\clip.py”; “\[chore\] pass pre-commit hook”; “!4 \[fix\] parse host and port for policy server from cli \ \[fix\] parse host and port for policy server from cli”; and 5 more.
scripts/eval · medium confidence · unverified
9 commits (1 fix) modifying internmanip/configs/model
A change to existing behaviour in internmanip/configs/model — 9 commits (1 fix), 4 files: “\[chore\] Optimized the train/eval data for genmanip, ensuring consistency with the IROS competition. \* Merge branch 'develop' into 'public' \* !53 \[feat\] add soft/hard task sr selection \* !52 \[feat\] optimizing parallel eval & update doc \* !51 update the genmanip data downloading \* !50 重新修改slice,支持default to None \* 更新文件: agent\cfg.py \ 更新文件: start\evaluator.py \ 更新文件: base.py \* 更新文件: dp\cfg.py \ 更新文件: dp\_clip.py”; “\[chore\] pass pre-commit hooks”; “\[chore\] pass pre-commit hook”; and 6 more.
internmanip/configs/model · medium confidence · unverified
9 commits (4 fixes) modifying internmanip/evaluator
A change to existing behaviour in internmanip/evaluator — 9 commits (4 fixs), 7 files: “\[chore\] Optimized the train/eval data for genmanip, ensuring consistency with the IROS competition. \* Merge branch 'develop' into 'public' \* !53 \[feat\] add soft/hard task sr selection \* !52 \[feat\] optimizing parallel eval & update doc \* !51 update the genmanip data downloading \* !50 重新修改slice,支持default to None \* 更新文件: agent\cfg.py \ 更新文件: start\evaluator.py \ 更新文件: base.py \* 更新文件: dp\cfg.py \ 更新文件: dp\_clip.py”; “\[chore\] pass pre-commit hook”; “\[refactor\] Refactor documentation, complete feature implementation, and add challenge setup”; and 6 more (refs \#22).
internmanip/evaluator · medium confidence · unverified
Fixes
6 commits (4 fixes) fixing challenge/bash\_scripts
A fix in challenge/bash\scripts — 6 commits (4 fixs), 1 file: “\[chore\] Optimized the train/eval data for genmanip, ensuring consistency with the IROS competition. \ Merge branch 'develop' into 'public' \* !53 \[feat\] add soft/hard task sr selection \* !52 \[feat\] optimizing parallel eval & update doc \* !51 update the genmanip data downloading \* !50 重新修改slice,支持default to None \* 更新文件: agent\cfg.py \ 更新文件: start\evaluator.py \ 更新文件: base.py \* 更新文件: dp\cfg.py \ 更新文件: dp\clip.py”; “!48 \[doc\] add quay tutorial in guidelines \ \[fix\] update eval.sh \* \[doc\] update guidelines \* \[doc\] update guidelines”; “!76 \[fix\] update to v0.1 \* update pi0 \* remove useless comment \* remove hard code \* remove hard code \* remove hard code in internmanip \* !75 update eval demo \* !74 \[fix\] fix genmanip eval bug by update InternUtopia \* !73 \[test\] added gr00t n1\5 test cases for genmanip training and testing \ !52 \[feat\] optimizing parallel eval & update doc”; and 3 more.
_challenge/bash\scripts · medium confidence · unverified
Test coverage
1 commit adding/updating tests in tests
Test coverage in tests — 1 commit (1 fix), 3 files: “!76 \[fix\] update to v0.1 \* update pi0 \* remove useless comment \* remove hard code \* remove hard code \* remove hard code in internmanip \* !75 update eval demo \* !74 \[fix\] fix genmanip eval bug by update InternUtopia \* !73 \[test\] added gr00t n1\5 test cases for genmanip training and testing \ !52 \[feat\] optimizing parallel eval & update doc”.
tests · high confidence · unverified
Written by cca.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.
How this codebase got here
Baseline
- First survey — no prior run to compare against. CAI 41.5.
Lenses
- Code Health 61.1
- Architecture 98.7
- Maturity 48.7
- Readiness 22.8
- Security 52.4
- Performance 100.0
Changes since last survey
- 59 commits — 38 feature/other, 21 fixes
By area
- (root) — 11 commits
- internmanip/model — 6 commits
- challenge/guidelines.md — 5 commits
- challenge/onsite_competition — 5 commits
- internmanip/benchmarks — 5 commits
- internmanip/agent — 4 commits
- run_configs/eval — 4 commits
- run_configs/train — 3 commits
- challenge/bash_scripts — 2 commits
- challenge/run_configs — 2 commits
- internmanip/configs — 2 commits
- internmanip/dataset — 2 commits
- scripts/eval — 2 commits
- assets/_static — 1 commit
- bash_scripts/internmanip_demo.sh — 1 commit
- internmanip/demo_data — 1 commit
- internmanip/unit_test_script — 1 commit
- internmanip/utils — 1 commit
- run_configs/examples — 1 commit
Notable commits
- fix: [doc] update guidelines & modify eval.sh * [doc] update challenge/guidelines.md * [fix] modify the executable permissions and content of eval.sh
- fix: !10 fix gr00t-n1.5 training and genmanip eval bugs * [fix] gr00t-n1.5 training and genmanip eval bugs
- fix: !2 fix bugs in close loop eval * remove hard code * rename get_action in gr00tn15 to inference * update gr00t eval config * update gr00t_n15 eval config
- fix: !4 [fix] parse host and port for policy server from cli * [fix] parse host and port for policy server from cli
- fix: !48 [doc] add quay tutorial in guidelines * [fix] update eval.sh * [doc] update guidelines * [doc] update guidelines
- fix: !5 verify and fix bugs for evaluating gr00t on simpler * verify and fix bugs for evaluating gr00t on simpler * verify and fix bugs for evaluating gr00t on simpler * remove manual model registry
- fix: !76 [fix] update to v0.1 * update pi0 * remove useless comment * remove hard code * remove hard code * remove hard code in internmanip * !75 update eval demo * !74 [fix] fix genmanip eval bug by update InternUtopia * !73 [test] added gr00t n1_5 test cases for genmanip training and testing * !52 [feat] optimizing parallel eval & update doc
- fix: !78 [fix] add tree in requirements * [fix] add tree in requirements
- fix: !80 [fix] fix data_config key and iros doc * [fix] fix data_config key and iros doc
- fix: Dev/fix agent structure
- fix: Fixed the issue where tune_llm and tune_vision had no effect
- fix: [chore] rebase develop to the public - [refactor] Refactor BaseAgent, AgentClient, AgentServer to disentangle model and env - [fix] fix bugs of genmanip agents, add challenge configs
- fix: [fix] add .gitmodules
- fix: [fix] fix hardcoded scripts * !58 [Fix] fix hardcoded gripper key in dataset augmentation * update files: gr00t_n1_5_on_genmanip.py
- fix: [fix] set gains for franka robotiq robot in gemanip & replace genmanip bench dataset repo id
- fix: [fix] specify transformers version in requirements (#5)
- fix: [fix] update doc
- fix: [fix] update robot usd and default result path * [fix] fix the default result path * [fix] replace aloha usd file
- fix: fix GenmanipAgent index mapping * [fix] GenmanipAgent index mapping
- fix: fix hard-coded name and paths
- …and 39 more
Written by cca.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.
Survey your own repository
InternRobotics/InternManip was measured the same way every project in this corpus was: the same rubric, at a pinned commit, with the result published in full. Point a surveyor at a repository you know and see whether you agree with it.
About this page
- The score is its most recent published measurement, taken on 16 November 2026 at a pinned commit. It is not a live figure and does not change until the project is measured again.
- Measured at commit 87dddb0fc0ba48e80551fdb16ad42faabe905ec2 — the exact code this score is about.
- Scored under rubric-2026.11.4 — the same rubric and the same method as every other entry in this index.
- Measured by Canine Code Analyzer (cca.canine.dev) using cca measure-2026.11.83+preprod-a489942cf94b.