diff --git a/harness-engineering-bench/CONFIGURATION.md b/harness-engineering-bench/CONFIGURATION.md
index e28dc749..1cbd1549 100644
--- a/harness-engineering-bench/CONFIGURATION.md
+++ b/harness-engineering-bench/CONFIGURATION.md
@@ -157,25 +157,25 @@ benchmark can be checked against the others at a glance.
 
 ## Per-benchmark values
 
-| | gaia | officeqa | swe-atlas-qna | tau3 | browsecomp-plus |
-|---|---|---|---|---|---|
-| target model | gpt-5.4-mini ◇ | deepseek-v4-flash | gpt-oss-120b | deepseek-v4-flash | deepseek-v4-flash |
-| held-out baseline (K=3) ◆ | 0.621 ±0.052 | 0.341 ±0.033 | 0.068 ±0.026 (agg 0.632) | 0.732 ±0.010 | 0.462 ±0.028 |
-| split dev/val/test | 33/66/66 | 49/98/99 | 25/49/50 | 75/150/150 | 33/66/66 |
-| dev budget (runs / cases) | 100 / 132 | 100 / 196 | 100 / 100 | 100 / 300 | 100 / 132 |
-| val budget (runs / cases) | 100 / 264 | 100 / 392 | 100 / 196 | 100 / 600 | 100 / 264 |
-| gateway max_tokens (evaluation, finalization each) ¶ | 2 B | 3 B | 2 B | 4 B | 2 B |
-| max_concurrency (cases in flight) § | 24 | 24 | 24 | 24 | 24 |
-| timeout_seconds (per eval) ‖ | 7200 | 28800 | 90000 | 79200 | 39600 |
-| case_timeout_seconds = declared † | 600 | 1800 | 10800 | 3600 | 3600 |
-| task_agent_timeout_seconds (declared) | 600 | 1800 | 10800 | 3600 | 3600 |
-| declared `[verifier] timeout_sec` | 300 | 300 | 900 | 300 | 300 |
-| declared `build_timeout_sec` | 300 | 600 | 600 | 600 | 7200 |
-| verifier_timeout_seconds ‖ | 14400 | 54000 | 176400 | 158400 | 75600 |
-| BASH_MAX_TIMEOUT_MS (tool) ¤ | 3600 s | 10800 s | 39600 s | 32400 s | 14400 s |
-| harness_user | harness | harness | null ‡ | null ‡ | null ‡ |
-| task_services_use_upstream | false | false | true (rubric judge) | true (user-sim + grader) | true (answer judge) |
-| task-specific extras | — | `--no-force-build` (prebuilt corpus image) | `keepalive` --ek (ENTRYPOINT images) | `TAU2_*` model pins | pinned 2.2 GB BM25 index |
+| | gaia | officeqa | swe-atlas-qna | tau3 | browsecomp-plus | swe-bench-pro |
+|---|---|---|---|---|---| --- |
+| target model | gpt-5.4-mini ◇ | deepseek-v4-flash | gpt-oss-120b | deepseek-v4-flash | deepseek-v4-flash | gpt-4o ◈ |
+| held-out baseline (K=3) ◆ | 0.621 ±0.052 | 0.341 ±0.033 | 0.068 ±0.026 (agg 0.632) | 0.732 ±0.010 | 0.462 ±0.028 | 0.294 ±0.008 ◈ |
+| split dev/val/test | 33/66/66 | 49/98/99 | 25/49/50 | 75/150/150 | 33/66/66 | 146/292/293 ◈ |
+| dev budget (runs / cases) | 100 / 132 | 100 / 196 | 100 / 100 | 100 / 300 | 100 / 132 | 100 / 146 ◈ |
+| val budget (runs / cases) | 100 / 264 | 100 / 392 | 100 / 196 | 100 / 600 | 100 / 264 | 100 / 292 ◈ |
+| gateway max_tokens (evaluation, finalization each) ¶ | 2 B | 3 B | 2 B | 4 B | 2 B | 100 M |
+| max_concurrency (cases in flight) § | 24 | 24 | 24 | 24 | 24 | 8 ◈ |
+| timeout_seconds (per eval) ‖ | 7200 | 28800 | 90000 | 79200 | 39600 | 28800 |
+| case_timeout_seconds = declared † | 600 | 1800 | 10800 | 3600 | 3600 | 1800 ◈ |
+| task_agent_timeout_seconds (declared) | 600 | 1800 | 10800 | 3600 | 3600 | 3000 |
+| declared `[verifier] timeout_sec` | 300 | 300 | 900 | 300 | 300 | n/a (registry dataset) |
+| declared `build_timeout_sec` | 300 | 600 | 600 | 600 | 7200 | n/a (registry dataset) |
+| verifier_timeout_seconds ‖ | 14400 | 54000 | 176400 | 158400 | 75600 | 28800 |
+| BASH_MAX_TIMEOUT_MS (tool) ¤ | 3600 s | 10800 s | 39600 s | 32400 s | 14400 s | n/a |
+| harness_user | harness | harness | null ‡ | null ‡ | null ‡ | harness |
+| task_services_use_upstream | false | false | true (rubric judge) | true (user-sim + grader) | true (answer judge) | false |
+| task-specific extras | — | `--no-force-build` (prebuilt corpus image) | `keepalive` --ek (ENTRYPOINT images) | `TAU2_*` model pins | pinned 2.2 GB BM25 index | registry dataset; `expose_case_resources: false`; sampled variant ◈ |
 
 ## Conventions
 
@@ -271,6 +271,52 @@ agent's reason/search-only-turn crash was fixed.
 
 **gaia and tau3 are too noisy for single-run comparisons.** gaia's own three rounds spanned 0.554-0.682 (sd 0.052), and tau3's optimizer scored one *unchanged* harness at 0.800 and 0.547 on development -- its user-simulator and NL-assertion grader are both LLMs, so their variance rides on every eval. Treat a gaia or tau3 delta under ~0.1 as unresolved. Their splits are not the problem: domain mix matches to the percentage point across all three partitions (airline 13%, banking 26%, retail 30%, telecom 31%), as does telecom persona difficulty.
 
+◈ swe-bench-pro is the newest benchmark and the only one with **two** configs.
+The column above describes the canonical `baseline/build.yaml` (full 731-instance
+dataset, split 146/292/293). Alongside it sits `baseline/build.sample.yaml`, a
+subsampled variant for optimizer runs; the canonical config is unchanged by it.
+
+*Why a variant.* The full split is unaffordable for an optimizer loop: each case
+builds a real repository and runs its test suite, so the held-out baseline alone
+cost ~7.8h of wall clock for three rounds, and an optimizer evaluates candidates
+many times over. The variant is 33/66/66, matching gaia and browsecomp-plus, on
+`fireworks_ai/deepseek-v4-flash`, `max_concurrency: 24`, 4x case budgets
+(132/264), and `case_timeout_seconds: 3000`.
+
+*The sample is nested.* The full split is computed first and each partition is
+then narrowed within itself (`partition_swe_bench_pro.py --sample`), so
+`partitions/sample/test.json` is a strict subset of `partitions/test.json`. No
+instance can migrate from test into development, and the baseline needed no
+re-measuring: **0.292 ±0.033** over the 66 sampled cases is the same K=3 trials
+that produced the full-partition 0.294 ±0.008, recomputed. The point estimate
+barely moves; the spread widens roughly fourfold because n falls 293 → 66, which
+matters when reading a candidate's margin over that floor.
+
+*Three known inconsistencies in the canonical config*, all pre-existing and left
+alone here rather than silently changed:
+
+- `case_timeout_seconds: 1800` against `task_agent_timeout_seconds: 3000` implies
+  a 0.6 multiplier, but the pinned 0.294 was measured at **1.0**: all 879 sampled
+  trial configs record `agent_timeout_multiplier=1.0`. The config value is wrong,
+  not the measurement. Every sibling has these two equal.
+- `model: gpt-4o`, but the 0.294 baseline was measured against
+  **qwen-3.6-27b**. So the recomputed 0.292 the variant pins is also a qwen
+  number while the variant targets deepseek; treat it as provisional until a K=3
+  deepseek pass over the 66 replaces it (~1.8h at `max_concurrency: 24`).
+- **The pinned 0.294 was measured on a different agent than the one in the tree**,
+  which is the same staleness the note above describes, in its strongest form. The
+  qwen-era agent was 630 lines on Chat Completions with an explicit local
+  `messages` list and edited files on 571 of 879 cases (write_file 1108,
+  apply_patch 899, submit 85). The committed seed is Responses-only, delegates its
+  conversation to `previous_response_id`, and edits nothing: over 66 sampled cases
+  it called write_file, apply_patch, and submit **zero** times and scored 0.0000.
+  So this benchmark is the one the re-pin above did not cover, and it needs
+  `scripts/rescore_candidate.py --seed` more than any of the five that did.
+
+Also unlike the siblings, swe-bench-pro's canonical config still has
+`score_baseline: true`, so it re-measures the seed every run instead of pinning
+it. The variant sets `false` and pins, as everything else does.
+
 ◇ gaia is the exception to the deepseek-v4-flash default: it is multimodal and
 that model is text-only. Verified against the same litellm endpoint the gateway
 proxies to — gpt-5.4-mini returns 200 for every request shape the gaia agent
@@ -320,7 +366,7 @@ per benchmark from its own case counts; see each `build.yaml` for the arithmetic
 the per-key bucket, so more keys buy proportionally more parallel runs:
 
 | counter | limit | scope |
-|---|---|---|
+|---|---|---| --- |
 | `x-ratelimit-api_key-limit-tokens` | 10 M TPM | **per key** |
 | `x-ratelimit-api_key-limit-requests` | 5 000 RPM | **per key** |
 
diff --git a/harness-engineering-bench/swe-bench-pro/baseline/build.sample.yaml b/harness-engineering-bench/swe-bench-pro/baseline/build.sample.yaml
new file mode 100644
index 00000000..4599e2c3
--- /dev/null
+++ b/harness-engineering-bench/swe-bench-pro/baseline/build.sample.yaml
@@ -0,0 +1,155 @@
+# Subsampled SWE-bench-Pro variant, for optimizer runs.
+#
+# This is a SECOND configuration alongside build.yaml, not a replacement. The
+# canonical build.yaml keeps the full 146/292/293 split and its held-out
+# baseline of 0.2940 +-0.0082; nothing here touches it.
+#
+# Why a variant exists: the full split is unaffordable for an optimizer loop.
+# Each case builds a real repository and runs its (often slow) test suite, so
+# the full-split held-out baseline alone cost ~7.8h of wall clock for three
+# rounds, and an optimizer evaluates candidates many times over. 33/66/66
+# matches gaia and browsecomp-plus, the suite's tightest split.
+#
+# The sampled partitions are NESTED inside the canonical ones: the full split is
+# computed first and each partition is then narrowed within itself, so
+# partitions/sample/test.json is a strict subset of partitions/test.json. No
+# instance can migrate from test into development, and the baseline below did
+# not need re-measuring.
+name: vero/optimize-swe-bench-pro-sample
+description: >-
+  Improve a code-editing SWE-bench-Pro agent while preserving the Harbor agent
+  interface. The target must edit a checked-out repository so the task's hidden
+  test suite passes; the task-source verifier runs the suite for the reward.
+agent_repo: target
+# SWE-bench-Pro ships as the `swebenchpro` dataset in the DEFAULT Harbor registry
+# (731 instances), not as an `<org>/<name>@sha256:<digest>` package like
+# swe-atlas-qna or tau3. Its tasks resolve to git-backed task ids under
+# laude-institute/harbor-datasets at commit c8e8f3fac7097accaacf261d74c3d6f441de45b1,
+# so the version pin is the registry version rather than a content digest.
+task_source: swebenchpro@1.0
+# Shared with the canonical split on purpose: the manifest pins all 731 refs and
+# the build loader reads only task names from it, so a second copy would just be
+# a second thing to keep in sync.
+task_manifest: ../partitions/manifest.json
+agent_import_path: swebench_pro_agent.agent:SweBenchProAgent
+harbor_requirement: harbor[modal]==0.20.0
+
+partition_files:
+  development: ../partitions/sample/development.json
+  validation: ../partitions/sample/validation.json
+  test: ../partitions/sample/test.json
+
+agent_access:
+  - partition: development
+    disclosure: full
+    # false, unlike the sibling benchmarks: VeRO materializes case resources via
+    # PackageDatasetClient, which only accepts `<org>/<name>` package refs. A
+    # registry dataset such as swebenchpro@1.0 cannot be exposed that way.
+    expose_case_resources: false
+    total_runs: 100
+    total_cases: 132  # 4x the partition, the suite-wide budget convention
+  - partition: validation
+    disclosure: aggregate
+    expose_case_resources: false
+    min_aggregate_cases: 5
+    total_runs: 100
+    total_cases: 264  # 4x the partition, the suite-wide budget convention
+
+selection_partition: validation
+targets:
+  - partition: test
+    reward_key: reward
+    # Held-out K=3 mean over the 66 sampled cases. Recomputed from the very same
+    # trials that produced the canonical full-partition 0.2940 +-0.0082, which is
+    # sound only because the sample is a strict subset; the recomputation
+    # reproduces 0.2940 +-0.0082 exactly on the full set first.
+    #
+    # CAVEAT: those trials targeted qwen-3.6-27b, whereas this variant targets
+    # deepseek-v4-flash. Treat 0.2923 as provisional until a K=3 pass on deepseek
+    # over these 66 cases replaces it (~1.8h at max_concurrency 24).
+    #
+    # The +-0.0332 is not a typo: it is 4x the full split's +-0.0082 because n
+    # falls 293 -> 66. Worth remembering when reading a candidate's margin.
+    baseline_reward: 0.2923
+    failure_value: 0.0
+    max_attempts: 1
+    # The held-out eval is noisy: score the selected candidate 3x per case and
+    # average, so the final reward is comparable to the pinned baseline, which was
+    # itself pooled over 3 rounds. Without this the candidate carries ~sqrt(3) more
+    # standard error than the floor it is judged against.
+    # Per-target override - search/validation keep the global n_attempts (1).
+    n_attempts: 3
+    aggregate_attempts: mean
+
+evaluation_set_name: swe-bench-pro-sample
+objective:
+  selector:
+    metric: score
+  direction: maximize
+reward_mode: submit  # agent picks; falls back to auto_best, then current version
+baseline_floor: false  # gates on validation while reward is on test; opt-in only
+score_baseline: false  # pinned as baseline_reward above, like every sibling
+rescore_top_k: 3
+rescore_attempts: 1
+
+model: fireworks_ai/deepseek-v4-flash
+environment_name: ${inner_env:-modal}
+# inner eval sandboxes share a dedicated Modal app instead of the __harbor__ default
+extra_harbor_args: ["--ek", "app_name=harness-engineering-bench", "--ek", "sandbox_idle_timeout_secs=3600"]
+harbor_python_version: "3.12"
+n_attempts: 1
+max_retries: 1
+infrastructure_max_attempts: 3
+infrastructure_retry_delay_seconds: 5
+aggregate_attempts: best
+feedback_transcripts: true
+feedback_max_bytes: 16000
+expose_attempt_detail: false
+# Timeouts are bumped well above the GAIA baseline: SWE-bench-Pro tasks build a
+# real repository and run its (often slow) test suite, so both the per-case and
+# the verifier budgets need substantially more headroom than a short-answer task.
+# Both mirror the task package's own `[agent] timeout_sec = 3000`, so the
+# multiplier is 1.0, matching every sibling benchmark. (The canonical build.yaml
+# still reads 1800 here, implying 0.6, but its published 0.2940 was in fact
+# measured at 1.0: all 879 sampled trial configs record
+# agent_timeout_multiplier=1.0. That file's value is wrong and worth a separate
+# one-line fix; this variant is simply spelled correctly from the start.)
+timeout_seconds: 28800
+case_timeout_seconds: 3000
+task_agent_timeout_seconds: 3000
+max_concurrency: 24  # matching the rest of the suite
+error_rate_threshold: 0.1
+verifier_timeout_seconds: 28800
+secrets:
+  - MODAL_TOKEN_ID
+  - MODAL_TOKEN_SECRET
+  - WANDB_API_KEY
+  - WANDB_BASE_URL  # self-hosted W&B (scaleai.wandb.io)
+
+# candidate harness runs as an unprivileged uid, unable to read held-out state.
+harness_user: harness
+
+wandb:
+  project: vero-swe-bench-pro
+  group: swe-bench-pro-sample  # distinct from the full-split group
+  log_traces: true
+
+inference_gateway:
+  upstream_api_key_env: OPENAI_API_KEY
+  upstream_base_url_env: OPENAI_BASE_URL
+  producer:
+    # NOTE: gpt-5.3-codex has no /responses deployment on the LiteLLM proxy
+    # (/chat/completions is 200, /responses is 404 DeploymentNotFound), and the
+    # codex agent is Responses-native, so launch with
+    # `--model gpt-5.6-sol` (or gpt-4.1) until that deployment exists.
+    allowed_models: ["${optimizer_model:-gpt-5.6-sol}"]
+    max_concurrency: 8
+  evaluation:
+    # Prefixed form only: the agents strip just `openai/` before calling the
+    # gateway, which then matches this allow-list as an exact string.
+    allowed_models: [fireworks_ai/deepseek-v4-flash]
+    max_requests: 15000
+    max_tokens: 100000000
+    max_concurrency: 64
+instruct_multifidelity: true
+instruct_exhaust_budget: true
diff --git a/harness-engineering-bench/swe-bench-pro/partitions/sample/development.json b/harness-engineering-bench/swe-bench-pro/partitions/sample/development.json
new file mode 100644
index 00000000..14439df1
--- /dev/null
+++ b/harness-engineering-bench/swe-bench-pro/partitions/sample/development.json
@@ -0,0 +1,35 @@
+[
+  "instance_ansible__ansible-12734fa21c08a0ce8c84e533abdc560db2eb1955-v7eee2454f617569fd6889f2211f75bc02a35f9f8",
+  "instance_ansible__ansible-1a4644ff15355fd696ac5b9d074a566a80fe7ca3-v30a923fb5c164d6cd18280c02422f75e611e8fb2",
+  "instance_ansible__ansible-34db57a47f875d11c4068567b9ec7ace174ec4cf-v1055803c3a812189a1133297f7f5468579283f86",
+  "instance_ansible__ansible-b8025ac160146319d2b875be3366b60c852dd35d-v0f01c69f1e2528b935359cfe578530722bca2c59",
+  "instance_ansible__ansible-fb144c44144f8bd3542e71f5db62b6d322c7bd85-vba6da65a0f3baefda7a058ebbd0a8dcafb8512f5",
+  "instance_element-hq__element-web-27139ca68eb075a4438c18fca184887002a4ffbc-vnan",
+  "instance_element-hq__element-web-56c7fc1948923b4b3f3507799e725ac16bcf8018-vnan",
+  "instance_flipt-io__flipt-8bd3604dc54b681f1f0f7dd52cbc70b3024184b6",
+  "instance_flipt-io__flipt-967855b429f749c28c112b8cb1b15bc79157f973",
+  "instance_flipt-io__flipt-b2cd6a6dd73ca91b519015fd5924fde8d17f3f06",
+  "instance_flipt-io__flipt-f1bc91a1b999656dbdb2495ccb57bf2105b84920",
+  "instance_future-architect__vuls-0ec945d0510cdebf92cdd8999f94610772689f14",
+  "instance_future-architect__vuls-7e91f5ef7e5712b1a3d7d5066ad6607e9debc21c",
+  "instance_future-architect__vuls-d576b6c6c15e56c47cc3e26f5878867677d4a9ea",
+  "instance_gravitational__teleport-0ecf31de0e98b272a6a2610abe1bbedd379a38a3-vce94f93ad1030e3136852817f2423c1b3ac37bc4",
+  "instance_gravitational__teleport-1a77b7945a022ab86858029d30ac7ad0d5239d00-vee9b09fb20c43af7e520f57e9239bbcf46b7113d",
+  "instance_gravitational__teleport-8302d467d160f869b77184e262adbe2fbc95d9ba-vce94f93ad1030e3136852817f2423c1b3ac37bc4",
+  "instance_internetarchive__openlibrary-630221ab686c64e75a2ce253c893c033e4814b2e-v93c53c13d5f9b383ebb411ee7750b49dcd1a34c6",
+  "instance_internetarchive__openlibrary-bdba0af0f6cbaca8b5fc3be2a3080f38156d9c92-ve8c8d62a2b60610a3c4631f5f23ed866bada9818",
+  "instance_internetarchive__openlibrary-d40ec88713dc95ea791b252f92d2f7b75e107440-v13642507b4fc1f8d234172bf8129942da2c2ca26",
+  "instance_internetarchive__openlibrary-d8162c226a9d576f094dc1830c4c1ffd0be2dd17-v76304ecdb3a5954fcf13feb710e8c40fcf24b73c",
+  "instance_navidrome__navidrome-29bc17acd71596ae92131aca728716baf5af9906",
+  "instance_navidrome__navidrome-8d56ec898e776e7e53e352cb9b25677975787ffc",
+  "instance_nodebb__nodebb-8ca65b0c78c67c1653487c02d1135e1b702185e1-vf2cf3cbd463b7ad942381f1c6d077626485a1e9e",
+  "instance_nodebb__nodebb-97c8569a798075c50e93e585ac741ab55cb7c28b-vf2cf3cbd463b7ad942381f1c6d077626485a1e9e",
+  "instance_protonmail__webclients-6e1873b06df6529a469599aa1d69d3b18f7d9d37",
+  "instance_protonmail__webclients-c5a2089ca2bfe9aa1d85a664b8ad87ef843a1c9c",
+  "instance_protonmail__webclients-fc9d535e9beb3ae30a52a7146398cadfd6e30606",
+  "instance_qutebrowser__qutebrowser-1a9e74bfaf9a9db2a510dc14572d33ded6040a57-v2ef375ac784985212b1805e1d0431dc8f1b3c171",
+  "instance_qutebrowser__qutebrowser-2e961080a85d660148937ee8f0f6b3445a8f2c01-v363c8a7e5ccdf6968fc7ab84a2053ac78036691d",
+  "instance_qutebrowser__qutebrowser-ebfe9b7aa0c4ba9d451f993e08955004aaec4345-v059c6fdc75567943479b23ebca7c07b5e9a7f34c",
+  "instance_qutebrowser__qutebrowser-ff1c025ad3210506fc76e1f604d8c8c27637d88e-v363c8a7e5ccdf6968fc7ab84a2053ac78036691d",
+  "instance_tutao__tutanota-db90ac26ab78addf72a8efaff3c7acc0fbd6d000-vbc0d9ba8f0071fbe982809910959a6ff8884dbbf"
+]
diff --git a/harness-engineering-bench/swe-bench-pro/partitions/sample/test.json b/harness-engineering-bench/swe-bench-pro/partitions/sample/test.json
new file mode 100644
index 00000000..b7291c28
--- /dev/null
+++ b/harness-engineering-bench/swe-bench-pro/partitions/sample/test.json
@@ -0,0 +1,68 @@
+[
+  "instance_ansible__ansible-4c5ce5a1a9e79a845aff4978cfeb72a0d4ecf7d6-v1055803c3a812189a1133297f7f5468579283f86",
+  "instance_ansible__ansible-709484969c8a4ffd74b839a673431a8c5caa6457-vba6da65a0f3baefda7a058ebbd0a8dcafb8512f5",
+  "instance_ansible__ansible-8127abbc298cabf04aaa89a478fc5e5e3432a6fc-v30a923fb5c164d6cd18280c02422f75e611e8fb2",
+  "instance_ansible__ansible-83fb24b923064d3576d473747ebbe62e4535c9e3-vba6da65a0f3baefda7a058ebbd0a8dcafb8512f5",
+  "instance_ansible__ansible-a6e671db25381ed111bbad0ab3e7d97366395d05-v0f01c69f1e2528b935359cfe578530722bca2c59",
+  "instance_ansible__ansible-d2f80991180337e2be23d6883064a67dcbaeb662-vba6da65a0f3baefda7a058ebbd0a8dcafb8512f5",
+  "instance_ansible__ansible-d62496fe416623e88b90139dc7917080cb04ce70-v0f01c69f1e2528b935359cfe578530722bca2c59",
+  "instance_ansible__ansible-de01db08d00c8d2438e1ba5989c313ba16a145b0-vba6da65a0f3baefda7a058ebbd0a8dcafb8512f5",
+  "instance_element-hq__element-web-71fe08ea0f159ccb707904d87f0a4aef205a167c-vnan",
+  "instance_element-hq__element-web-9bf77963ee5e036d54b2a3ca202fbf6378464a5e-vnan",
+  "instance_element-hq__element-web-ca8b1b04effb4fec0e1dd3de8e3198eeb364d50e-vnan",
+  "instance_element-hq__element-web-d06cf09bf0b3d4a0fbe6bd32e4115caea2083168-vnan",
+  "instance_element-hq__element-web-d405160080bbe804f7e9294067d004a7d4dad9d6-vnan",
+  "instance_flipt-io__flipt-2eac0df47b5ecc8bb05002d80383ceb08ab3620a",
+  "instance_flipt-io__flipt-507170da0f7f4da330f6732bffdf11c4df7fc192",
+  "instance_flipt-io__flipt-524f277313606f8cd29b299617d6565c01642e15",
+  "instance_flipt-io__flipt-56a620b8fc9ef7a0819b47709aa541cdfdbba00b",
+  "instance_flipt-io__flipt-b433bd05ce405837804693bebd5f4b88d87133c8",
+  "instance_flipt-io__flipt-c154dd1a3590954dfd3b901555fc6267f646a289",
+  "instance_flipt-io__flipt-d966559200183b713cdf3ea5007a7e0ba86a5afb",
+  "instance_flipt-io__flipt-dae029cba7cdb98dfb1a6b416c00d324241e6063",
+  "instance_future-architect__vuls-01441351c3407abfc21c48a38e28828e1b504e0c",
+  "instance_future-architect__vuls-54e73c2f5466ef5daec3fb30922b9ac654e4ed25",
+  "instance_future-architect__vuls-5af1a227339e46c7abf3f2815e4c636a0c01098e",
+  "instance_future-architect__vuls-61c39637f2f3809e1b5dad05f0c57c799dce1587",
+  "instance_future-architect__vuls-be7b9114cc9545e68fb0ee7bc63d7ec53d1a00ad",
+  "instance_future-architect__vuls-f0b3a8b1db98eb1bd32685f1c36c41a99c3452ed",
+  "instance_gravitational__teleport-3fa6904377c006497169945428e8197158667910-v626ec2a48416b10a88641359a169d99e935ff037",
+  "instance_gravitational__teleport-53814a2d600ccd74c1e9810a567563432b98386e-vce94f93ad1030e3136852817f2423c1b3ac37bc4",
+  "instance_gravitational__teleport-59d39dee5a8a66e5b8a18a9085a199d369b1fba8-v626ec2a48416b10a88641359a169d99e935ff037",
+  "instance_gravitational__teleport-89f0432ad5dc70f1f6a30ec3a8363d548371a718",
+  "instance_gravitational__teleport-db89206db6c2969266e664c7c0fb51b70e958b64",
+  "instance_gravitational__teleport-eefac60a350930e5f295f94a2d55b94c1988c04e-vee9b09fb20c43af7e520f57e9239bbcf46b7113d",
+  "instance_gravitational__teleport-f432a71a13e698b6e1c4672a2e9e9c1f32d35c12",
+  "instance_internetarchive__openlibrary-1351c59fd43689753de1fca32c78d539a116ffc1-v29f82c9cf21d57b242f8d8b0e541525d259e2d63",
+  "instance_internetarchive__openlibrary-30bc73a1395fba2300087c7f307e54bb5372b60a-v76304ecdb3a5954fcf13feb710e8c40fcf24b73c",
+  "instance_internetarchive__openlibrary-6a117fab6c963b74dc1ba907d838e74f76d34a4b-v13642507b4fc1f8d234172bf8129942da2c2ca26",
+  "instance_internetarchive__openlibrary-798055d1a19b8fa0983153b709f460be97e33064-v13642507b4fc1f8d234172bf8129942da2c2ca26",
+  "instance_internetarchive__openlibrary-91efee627df01e32007abf2d6ebf73f9d9053076-vbee42ad1b72fb23c6a1c874868a720b370983ed2",
+  "instance_internetarchive__openlibrary-dbbd9d539c6d4fd45d5be9662aa19b6d664b5137-v08d8e8889ec945ab821fb156c04c7d2e2810debb",
+  "instance_internetarchive__openlibrary-de6ae10512f1b5ef585c8341b451bc49c9fd4996-vfa6ff903cb27f336e17654595dd900fa943dcd91",
+  "instance_internetarchive__openlibrary-e1e502986a3b003899a8347ac8a7ff7b08cbfc39-v08d8e8889ec945ab821fb156c04c7d2e2810debb",
+  "instance_navidrome__navidrome-0488fb92cb02a82924fb1181bf1642f2e87096db",
+  "instance_navidrome__navidrome-66b74c81f115c78cb69910b0472eeb376750efc4",
+  "instance_navidrome__navidrome-d21932bd1b2379b0ebca2d19e5d8bae91040268a",
+  "instance_navidrome__navidrome-d613b1930688422122796b43acb3caf2538c8fd1",
+  "instance_navidrome__navidrome-f7d4fcdcc1a59d1b4f835519efb402897757e371",
+  "instance_nodebb__nodebb-0c81642997ea1d827dbd02c311db9d4976112cd4-vf2cf3cbd463b7ad942381f1c6d077626485a1e9e",
+  "instance_nodebb__nodebb-84e065752f6d7fbe5c08cbf50cb173ffb866b8fa-vf2cf3cbd463b7ad942381f1c6d077626485a1e9e",
+  "instance_nodebb__nodebb-bad15643013ca15affe408b75eba9e47cc604bb2-vd59a5728dfc977f44533186ace531248c2917516",
+  "instance_nodebb__nodebb-eb49a64974ca844bca061744fb3383f5d13b02ad-vnan",
+  "instance_protonmail__webclients-8142704f447df6e108d53cab25451c8a94976b92",
+  "instance_protonmail__webclients-b9387af4cdf79c2cb2a221dea33d665ef789512e",
+  "instance_protonmail__webclients-c6f65d205c401350a226bb005f42fac1754b0b5b",
+  "instance_protonmail__webclients-c8117f446c3d1d7e117adc6e0e46b0ece9b0b90e",
+  "instance_protonmail__webclients-cba6ebbd0707caa524ffee51c62b197f6122c902",
+  "instance_protonmail__webclients-da91f084c0f532d9cc8ca385a701274d598057b8",
+  "instance_qutebrowser__qutebrowser-0b621cb0ce2b54d3f93d8d41d8ff4257888a87e5-v2ef375ac784985212b1805e1d0431dc8f1b3c171",
+  "instance_qutebrowser__qutebrowser-21b426b6a20ec1cc5ecad770730641750699757b-v363c8a7e5ccdf6968fc7ab84a2053ac78036691d",
+  "instance_qutebrowser__qutebrowser-305e7c96d5e2fdb3b248b27dfb21042fb2b7e0b8-v2ef375ac784985212b1805e1d0431dc8f1b3c171",
+  "instance_qutebrowser__qutebrowser-479aa075ac79dc975e2e949e188a328e95bf78ff-vc2f56a753b62a190ddb23cd330c257b9cf560d12",
+  "instance_qutebrowser__qutebrowser-70248f256f93ed9b1984494d0a1a919ddd774892-v2ef375ac784985212b1805e1d0431dc8f1b3c171",
+  "instance_qutebrowser__qutebrowser-77c3557995704a683cdb67e2a3055f7547fa22c3-v363c8a7e5ccdf6968fc7ab84a2053ac78036691d",
+  "instance_qutebrowser__qutebrowser-8d05f0282a271bfd45e614238bd1b555c58b3fc1-v35616345bb8052ea303186706cec663146f0f184",
+  "instance_tutao__tutanota-fb32e5f9d9fc152a00144d56dd0af01760a2d4dc-vc4e41fd0029957297843cb9dec4a25c7c756f029",
+  "instance_tutao__tutanota-fbdb72a2bd39b05131ff905780d9d4a2a074de26-vbc0d9ba8f0071fbe982809910959a6ff8884dbbf"
+]
diff --git a/harness-engineering-bench/swe-bench-pro/partitions/sample/validation.json b/harness-engineering-bench/swe-bench-pro/partitions/sample/validation.json
new file mode 100644
index 00000000..647fbf20
--- /dev/null
+++ b/harness-engineering-bench/swe-bench-pro/partitions/sample/validation.json
@@ -0,0 +1,68 @@
+[
+  "instance_ansible__ansible-11c1777d56664b1acb56b387a1ad6aeadef1391d-v0f01c69f1e2528b935359cfe578530722bca2c59",
+  "instance_ansible__ansible-1b70260d5aa2f6c9782fd2b848e8d16566e50d85-vba6da65a0f3baefda7a058ebbd0a8dcafb8512f5",
+  "instance_ansible__ansible-1ee70fc272aff6bf3415357c6e13c5de5b928d9b-v1055803c3a812189a1133297f7f5468579283f86",
+  "instance_ansible__ansible-5f4e332e3762999d94af27746db29ff1729252c1-v0f01c69f1e2528b935359cfe578530722bca2c59",
+  "instance_ansible__ansible-77658704217d5f166404fc67997203c25381cb6e-v390e508d27db7a51eece36bb6d9698b63a5b638a",
+  "instance_ansible__ansible-9142be2f6cabbe6597c9254c5bb9186d17036d55-v0f01c69f1e2528b935359cfe578530722bca2c59",
+  "instance_ansible__ansible-d6d2251929c84c3aa883bad7db0f19cc9ff0339e-v30a923fb5c164d6cd18280c02422f75e611e8fb2",
+  "instance_ansible__ansible-e64c6c1ca50d7d26a8e7747d8eb87642e767cd74-v0f01c69f1e2528b935359cfe578530722bca2c59",
+  "instance_element-hq__element-web-1077729a19c0ce902e713cf6fab42c91fb7907f1-vnan",
+  "instance_element-hq__element-web-459df4583e01e4744a52d45446e34183385442d6-vnan",
+  "instance_element-hq__element-web-ad26925bb6628260cfe0fcf90ec0a8cba381f4a4-vnan",
+  "instance_element-hq__element-web-aec454dd6feeb93000380523cbb0b3681c0275fd-vnan",
+  "instance_element-hq__element-web-dae13ac8522fc6d41e64d1ac6e3174486fdcce0c-vnan",
+  "instance_flipt-io__flipt-0fd09def402258834b9d6c0eaa6d3b4ab93b4446",
+  "instance_flipt-io__flipt-40007b9d97e3862bcef8c20ae6c87b22ea0627f0",
+  "instance_flipt-io__flipt-86906cbfc3a5d3629a583f98e6301142f5f14bdb-v6bea0cc3a6fc532d7da914314f2944fc1cd04dee",
+  "instance_flipt-io__flipt-9f8127f225a86245fa35dca4885c2daef824ee55",
+  "instance_flipt-io__flipt-c12967bc73fdf02054cf3ef8498c05e25f0a18c0",
+  "instance_flipt-io__flipt-c1728053367c753688f114ec26e703c8fdeda125",
+  "instance_flipt-io__flipt-c6a7b1fd933e763b1675281b30077e161fa115a1",
+  "instance_flipt-io__flipt-c8d71ad7ea98d97546f01cce4ccb451dbcf37d3b",
+  "instance_future-architect__vuls-36456cb151894964ba1683ce7da5c35ada789970",
+  "instance_future-architect__vuls-4a72295de7b91faa59d90a5bee91535bbe76755d",
+  "instance_future-architect__vuls-50580f6e98eeb36f53f27222f7f4fdfea0b21e8d",
+  "instance_future-architect__vuls-ad2edbb8448e2c41a097f1c0b52696c0f6c5924d",
+  "instance_future-architect__vuls-e52fa8d6ed1d23e36f2a86e5d3efe9aa057a1b0d",
+  "instance_future-architect__vuls-ef2be3d6ea4c0a13674aaab08b182eca4e2b9a17-v264a82e2f4818e30f5a25e4da53b27ba119f62b5",
+  "instance_gravitational__teleport-10123c046e21e1826098e485a4c2212865a49d9f",
+  "instance_gravitational__teleport-3a5c1e26394df2cb4fb3f01147fb9979662972c5-vee9b09fb20c43af7e520f57e9239bbcf46b7113d",
+  "instance_gravitational__teleport-73cc189b0e9636d418c4470ecce0d9af5dae2f02-vee9b09fb20c43af7e520f57e9239bbcf46b7113d",
+  "instance_gravitational__teleport-a95b3ae0667f9e4b2404bf61f51113e6d83f01cd",
+  "instance_gravitational__teleport-b4e7cd3a5e246736d3fe8d6886af55030b232277",
+  "instance_gravitational__teleport-d873ea4fa67d3132eccba39213c1ca2f52064dcc-vce94f93ad1030e3136852817f2423c1b3ac37bc4",
+  "instance_gravitational__teleport-e6895d8934f6e484341034869901145fbc025e72-vce94f93ad1030e3136852817f2423c1b3ac37bc4",
+  "instance_internetarchive__openlibrary-09865f5fb549694d969f0a8e49b9d204ef1853ca-ve8c8d62a2b60610a3c4631f5f23ed866bada9818",
+  "instance_internetarchive__openlibrary-1be7de788a444f6255e89c10ef6aa608550604a8-v29f82c9cf21d57b242f8d8b0e541525d259e2d63",
+  "instance_internetarchive__openlibrary-2abe28b472ffed563a87cfe83685b161b35263b0-v13642507b4fc1f8d234172bf8129942da2c2ca26",
+  "instance_internetarchive__openlibrary-427f1f4eddfc54735ca451779d4f95bf683d1b0e-v0f5aece3601a5b4419f7ccec1dbda2071be28ee4",
+  "instance_internetarchive__openlibrary-89e4b4431fe7506c365a6f6eb6f6d048d04c044c-v08d8e8889ec945ab821fb156c04c7d2e2810debb",
+  "instance_internetarchive__openlibrary-9c392b60e2c6fa1d68cb68084b4b4ff04d0cb35c-v2d9a6c849c60ed19fd0858ce9e40b7cc8e097e59",
+  "instance_internetarchive__openlibrary-9cd47f4dc21e273320d9e30d889c864f8cb20ccf-v0f5aece3601a5b4419f7ccec1dbda2071be28ee4",
+  "instance_internetarchive__openlibrary-c506c1b0b678892af5cb22c1c1dbc35d96787a0a-v0f5aece3601a5b4419f7ccec1dbda2071be28ee4",
+  "instance_navidrome__navidrome-09ae41a2da66264c60ef307882362d2e2d8d8b89",
+  "instance_navidrome__navidrome-0a650de357babdcc8ce910fe37fee84acf4ed2fe",
+  "instance_navidrome__navidrome-3977ef6e0f287f598b6e4009876239d6f13b686d",
+  "instance_navidrome__navidrome-69e0a266f48bae24a11312e9efbe495a337e4c84",
+  "instance_navidrome__navidrome-e12a14a87d392ac70ee4cc8079e3c3e0103dbcb2",
+  "instance_nodebb__nodebb-3c85b944e30a0ba8b3ec9e1f441c74f383625a15-v4fbcfae8b15e4ce5d132c408bca69ebb9cf146ed",
+  "instance_nodebb__nodebb-51d8f3b195bddb13a13ddc0de110722774d9bb1b-vf2cf3cbd463b7ad942381f1c6d077626485a1e9e",
+  "instance_nodebb__nodebb-b1f9ad5534bb3a44dab5364f659876a4b7fe34c1-vnan",
+  "instance_nodebb__nodebb-f083cd559d69c16481376868c8da65172729c0ca-vnan",
+  "instance_protonmail__webclients-01b519cd49e6a24d9a05d2eb97f54e420740072e",
+  "instance_protonmail__webclients-01ea5214d11e0df8b7170d91bafd34f23cb0f2b1",
+  "instance_protonmail__webclients-6dcf0d0b0f7965ad94be3f84971afeb437f25b02",
+  "instance_protonmail__webclients-708ed4a299711f0fa79a907cc5847cfd39c0fc71",
+  "instance_protonmail__webclients-815695401137dac2975400fc610149a16db8214b",
+  "instance_protonmail__webclients-944adbfe06644be0789f59b78395bdd8567d8547",
+  "instance_qutebrowser__qutebrowser-1af602b258b97aaba69d2585ed499d95e2303ac2-v2ef375ac784985212b1805e1d0431dc8f1b3c171",
+  "instance_qutebrowser__qutebrowser-394bfaed6544c952c6b3463751abab3176ad4997-vafb3e8e01b31319c66c4e666b8a3b1d8ba55db24",
+  "instance_qutebrowser__qutebrowser-54bcdc1eefa86cc20790973d6997b60c3bba884c-v2ef375ac784985212b1805e1d0431dc8f1b3c171",
+  "instance_qutebrowser__qutebrowser-8f46ba3f6dc7b18375f7aa63c48a1fe461190430-v2ef375ac784985212b1805e1d0431dc8f1b3c171",
+  "instance_qutebrowser__qutebrowser-bedc9f7fadf93f83d8dee95feeecb9922b6f063f-v2ef375ac784985212b1805e1d0431dc8f1b3c171",
+  "instance_qutebrowser__qutebrowser-c09e1439f145c66ee3af574386e277dd2388d094-v2ef375ac784985212b1805e1d0431dc8f1b3c171",
+  "instance_qutebrowser__qutebrowser-ec2dcfce9eee9f808efc17a1b99e227fc4421dea-v5149fcda2a9a6fe1d35dfed1bade1444a11ef271",
+  "instance_tutao__tutanota-219bc8f05d7b980e038bc1524cb021bf56397a1b-vee878bb72091875e912c52fc32bc60ec3760227b",
+  "instance_tutao__tutanota-f3ffe17af6e8ab007e8d461355057ad237846d9d-vbc0d9ba8f0071fbe982809910959a6ff8884dbbf"
+]
diff --git a/harness-engineering-bench/swe-bench-pro/scripts/partition_swe_bench_pro.py b/harness-engineering-bench/swe-bench-pro/scripts/partition_swe_bench_pro.py
index 9be9ab02..89e13028 100644
--- a/harness-engineering-bench/swe-bench-pro/scripts/partition_swe_bench_pro.py
+++ b/harness-engineering-bench/swe-bench-pro/scripts/partition_swe_bench_pro.py
@@ -41,14 +41,33 @@
 TOTAL_TASKS = 731
 TARGET_COUNTS = {"development": 146, "validation": 292, "test": 293}
 
+# `--sample` writes a second, smaller split into partitions/sample/ for the
+# optimizer variant (baseline/build.sample.yaml). The canonical split above is
+# never touched by it. 33/66/66 matches gaia and browsecomp-plus, the suite's
+# tightest split, which is the right reference for its most expensive benchmark:
+# a case here builds a real repository and runs its test suite, so the full-split
+# held-out baseline alone cost ~7.8h of wall clock for three rounds.
+SAMPLE_COUNTS = {"development": 33, "validation": 66, "test": 66}
+
 
 def _parse_args() -> argparse.Namespace:
     parser = argparse.ArgumentParser()
     parser.add_argument(
         "--tasks-dir",
         type=Path,
-        required=True,
-        help="Directory containing the exported SWE-bench-Pro task directories",
+        help=(
+            "Directory containing the exported SWE-bench-Pro task directories. "
+            "Omit to re-derive the split from the committed manifest, which "
+            "already records every instance's name and stratum."
+        ),
+    )
+    parser.add_argument(
+        "--sample",
+        action="store_true",
+        help=(
+            "Write the smaller SAMPLE_COUNTS split to partitions/sample/ for the "
+            "optimizer variant. Leaves the canonical split untouched."
+        ),
     )
     parser.add_argument(
         "--output-dir",
@@ -97,6 +116,27 @@ def _read_tasks(path: Path) -> list[dict[str, str]]:
     return tasks
 
 
+def _tasks_from_manifest(output_dir: Path) -> list[dict[str, str]]:
+    """Re-read the (name, repository) pairs the committed manifest already pins.
+
+    Deriving the sample should not require re-exporting 731 task directories:
+    the manifest records every instance and its stratum verbatim, which is the
+    entire input `_allocate` needs. `--tasks-dir` remains the path for re-pinning
+    against a moved dataset.
+    """
+    manifest_path = output_dir / "manifest.json"
+    if not manifest_path.is_file():
+        raise ValueError("no existing manifest; pass --tasks-dir")
+    manifest = json.loads(manifest_path.read_text(encoding="utf-8"))
+    tasks = [
+        {"name": item["name"], "repository": item["repository"]}
+        for item in manifest["tasks"]
+    ]
+    if len(tasks) != TOTAL_TASKS:
+        raise ValueError(f"expected {TOTAL_TASKS} tasks, manifest has {len(tasks)}")
+    return tasks
+
+
 async def _fetch_registry_refs() -> tuple[str, dict[str, str]]:
     try:
         from harbor.registry.client.factory import RegistryClientFactory
@@ -187,6 +227,63 @@ def _allocate(tasks: list[dict[str, str]]) -> dict[str, list[str]]:
     return result
 
 
+def _subsample(
+    full: dict[str, list[str]], tasks: list[dict[str, str]]
+) -> dict[str, list[str]]:
+    """Shrink each partition to SAMPLE_COUNTS without letting tasks cross partitions.
+
+    Sampling happens strictly inside a partition, so every sampled set is a
+    subset of the corresponding full set. Two properties follow, and both are
+    why it is done this way rather than by re-running the split at a smaller
+    size: an instance can never move from test into development (a flat
+    re-split does move them, which would silently destroy the held-out
+    guarantee), and a reward already measured over the full partition can be
+    recomputed over the sample from the existing per-case trials.
+
+    Within a partition the same repository stratification and sha256 ordering
+    apply, so the sample keeps the partition's repo mix and stays reproducible.
+    """
+    repository = {task["name"]: task["repository"] for task in tasks}
+    sampled: dict[str, list[str]] = {}
+    for partition in PARTITIONS:
+        target = SAMPLE_COUNTS[partition]
+        members = full[partition]
+        if target > len(members):
+            raise ValueError(
+                f"{partition} sample of {target} exceeds its {len(members)} tasks"
+            )
+        strata: dict[str, list[str]] = defaultdict(list)
+        for name in members:
+            strata[repository[name]].append(name)
+
+        quota = {
+            stratum: target * len(names) // len(members)
+            for stratum, names in strata.items()
+        }
+        remainders = sorted(
+            strata,
+            key=lambda stratum: (
+                -(target * len(strata[stratum]) % len(members)),
+                stratum,
+            ),
+        )
+        cursor = 0
+        while sum(quota.values()) < target:
+            stratum = remainders[cursor % len(remainders)]
+            if quota[stratum] < len(strata[stratum]):
+                quota[stratum] += 1
+            cursor += 1
+
+        chosen: list[str] = []
+        for stratum in sorted(strata):
+            chosen.extend(sorted(strata[stratum], key=_stable_key)[: quota[stratum]])
+        chosen.sort()
+        if len(chosen) != target:
+            raise RuntimeError(f"wrong sampled {partition} size: {len(chosen)}")
+        sampled[partition] = chosen
+    return sampled
+
+
 def _render(
     tasks: list[dict[str, str]],
     partitions: dict[str, list[str]],
@@ -235,15 +332,37 @@ def _render(
     return rendered
 
 
+def _render_sample(partitions: dict[str, list[str]]) -> dict[str, str]:
+    """Just the three partition files. The manifest is shared with the full split.
+
+    Nothing benchmark-level changes for the sample: the same manifest pins the
+    same 731 refs, and the build loader reads only task names out of it. Writing
+    a second manifest would duplicate that pinning and let the two drift.
+    """
+    return {
+        f"{partition}.json": json.dumps(partitions[partition], indent=2) + "\n"
+        for partition in PARTITIONS
+    }
+
+
 def main() -> None:
     args = _parse_args()
-    tasks = _read_tasks(args.tasks_dir)
     args.output_dir = args.output_dir.expanduser().resolve()
-    if args.fetch_registry:
-        dataset_version, refs = asyncio.run(_fetch_registry_refs())
+    tasks = (
+        _read_tasks(args.tasks_dir)
+        if args.tasks_dir is not None
+        else _tasks_from_manifest(args.output_dir)
+    )
+    if args.sample:
+        # Derived from the canonical split, then narrowed inside each partition.
+        rendered = _render_sample(_subsample(_allocate(tasks), tasks))
+        args.output_dir = args.output_dir / "sample"
     else:
-        dataset_version, refs = _existing_refs(args.output_dir)
-    rendered = _render(tasks, _allocate(tasks), dataset_version, refs)
+        if args.fetch_registry:
+            dataset_version, refs = asyncio.run(_fetch_registry_refs())
+        else:
+            dataset_version, refs = _existing_refs(args.output_dir)
+        rendered = _render(tasks, _allocate(tasks), dataset_version, refs)
 
     if args.check:
         changed = [