SWE-bench-Live
Collection
The datasets for benchmarking and training LLM coding agents. • 3 items • Updated • 1
repo string | pull_number string | instance_id string | issue_numbers list | base_commit string | patch string | test_patch string | problem_statement string | hints_text string | all_hints_text string | commit_urls list | created_at string | commit_url string | rebuild_cmds list | test_cmds list | print_cmds list | log_parser string | FAIL_TO_PASS list | PASS_TO_PASS list | docker_image string |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
samtools/samtools | 2235 | samtools__samtools-2235 | [
"1563"
] | abaa1583614a12cb446e8927b92723fdfde2025c | diff --git a/coverage.c b/coverage.c
index 92f497dc8..7fcb34f7e 100644
--- a/coverage.c
+++ b/coverage.c
@@ -1,7 +1,7 @@
/* coverage.c -- samtools coverage subcommand
Copyright (C) 2018,2019 Florian Breitwieser
- Portions copyright (C) 2019-2021, 2023-2024 Genome Research Ltd.
+ Portions copyright (C) 201... | diff --git a/test/coverage/1.expected b/test/coverage/1.expected
new file mode 100644
index 000000000..35cd07c2a
--- /dev/null
+++ b/test/coverage/1.expected
@@ -0,0 +1,3 @@
+#rname startpos endpos numreads covbases coverage meandepth meanbaseq meanmapq
+T1 1 40 6 19 47.5 0.725 7.62 36.7
+T2 1 40 6 8 20 0.4 7.12 49
dif... | Minimum depth threshold for "samtools coverage"?
#### Is your feature request related to a problem? Please specify.
I'd like a simple way to calculate the coverage stats for a bam file. The closest out-of-the-box thing I can find is `samtools coverage`. I run the command like so:
```
samtools coverage input.bam
... | Looks like a good idea, we'll consider it.
Hi Did you implement it in Samtools?
It doesn't look it, but this line is the thing you could manually tweak in a local copy:
https://github.com/samtools/samtools/blob/develop/coverage.c#L603
If changed to `>=10` for example it'd need a minimum of 10 deep.
However loo... | Looks like a good idea, we'll consider it.
Hi Did you implement it in Samtools?
It doesn't look it, but this line is the thing you could manually tweak in a local copy:
https://github.com/samtools/samtools/blob/develop/coverage.c#L603
If changed to `>=10` for example it'd need a minimum of 10 deep.
However loo... | [
"https://github.com/samtools/samtools/commit/3a769cfa9a71df943b00068566a171269e2789d3",
"https://github.com/samtools/samtools/commit/6a807af098c28ef534b684aaa2b305e5776145d2"
] | 2025-07-10T19:03:03Z | https://github.com/samtools/samtools/tree/abaa1583614a12cb446e8927b92723fdfde2025c | [
"autoheader && autoconf -Wno-syntax ; ./configure --with-htslib=htslib ; make"
] | [
"make test V=1 2>&1 | tee test-output.log"
] | [
"cat test-output.log"
] | def parser(log: str) -> dict[str, str]:
import re
results = {}
# Match lines like: test_name: <command> .. ok / .. failed
test_re = re.compile(r'^(?P<name>[^\s].*?):\s*$')
status_re = re.compile(r'^\.\. (ok|failed)', re.IGNORECASE)
current_test = None
for line in log.splitlines():
m... | [
"test_coverage"
] | [
"test_large_positions",
"test_merge",
"test_collate",
"test_reference",
"test_split",
"test_quickcheck",
"test_markdup",
"test_bedcov",
"test_reheader",
"test_addrprg",
"test_sort",
"test_dict",
"test_head",
"test_fixmate",
"test_stats",
"test_idxstat",
"test_ampliconclip",
"test_i... | starryzhang/sweb.eval.x86_64.samtools_1776_samtools-2235 |
micropython/micropython | 17683 | micropython__micropython-17683 | [
"17657"
] | 554f114f181ee942ee3c74e44cef653604abbaef | diff --git a/ports/webassembly/objpyproxy.js b/ports/webassembly/objpyproxy.js
index 0eafd0dec53de..64703d78a5589 100644
--- a/ports/webassembly/objpyproxy.js
+++ b/ports/webassembly/objpyproxy.js
@@ -165,34 +165,35 @@ const py_proxy_handler = {
if (prop === "_ref") {
return target._ref;
... | diff --git a/tests/ports/webassembly/py_proxy_get.mjs b/tests/ports/webassembly/py_proxy_get.mjs
new file mode 100644
index 0000000000000..825de7cabeb78
--- /dev/null
+++ b/tests/ports/webassembly/py_proxy_get.mjs
@@ -0,0 +1,14 @@
+// Test `<py-obj> get <attr>` on the JavaScript side, which tests PyProxy.get.
+
+const ... | webassembly: getting JS symbols implicitly also throws errors
### Port, board and/or hardware
webassembly
### MicroPython version
MicroPython webassembly latest from *npm*
### Reproduction
In here we solved the `in` operation check https://github.com/micropython/micropython/pull/17604 but if any libraries tries to... | thanks @dpgeorge , any chance this can make it to npm too?
| [
"https://github.com/micropython/micropython/commit/c72a3e528d7909c212596b52de5f9a5fe0161f17"
] | 2025-07-15T08:45:08Z | https://github.com/micropython/micropython/tree/554f114f181ee942ee3c74e44cef653604abbaef | [
"make -C mpy-cross ; make -C ports/unix"
] | [
"make -C ports/unix test 2>&1 | tee test-output.log"
] | [
"cat test-output.log"
] | def parser(log: str) -> dict[str, str]:
import re
results: dict[str, str] = {}
for line in log.splitlines():
m = re.match(r'(pass|FAIL|skip)\s+(\S+)', line)
if m:
status_word, test_name = m.groups()
status_word = status_word.lower()
if status_word == "pass... | [
"micropython/heapalloc_fail_dict.py",
"micropython/heapalloc_int_from_bytes.py",
"float/float_divmod.py",
"extmod/json_load.py",
"extmod/re1.py",
"float/cmath_dunder.py",
"extmod/vfs_userfs.py",
"float/builtin_float_minmax.py",
"float/lexer.py",
"basics/struct1.py",
"basics/try_else.py",
"micr... | [
"basics/memoryview1.py",
"basics/int_big_rshift.py",
"io/file_stdio.py",
"extmod/re_sub.py",
"basics/bytearray_construct_endian.py",
"basics/dict_fromkeys.py",
"micropython/viper_binop_arith_uint.py",
"misc/non_compliant_lexer.py",
"basics/int_big_add.py",
"io/file_stdio2.py",
"basics/class_dict... | starryzhang/sweb.eval.x86_64.micropython_1776_micropython-17683 | |
timescale/timescaledb | 8559 | timescale__timescaledb-8559 | [
"6902"
] | bbbebbfa978465bfdf6d0bac418ee4d269ad4da8 | diff --git a/.unreleased/pr_8559 b/.unreleased/pr_8559
new file mode 100644
index 00000000000..79123483309
--- /dev/null
+++ b/.unreleased/pr_8559
@@ -0,0 +1,2 @@
+Fixes: #8559 Fix `timestamp out of range` using `end_offset=NULL` on CAgg refresh policy
+Thanks: @nofalx for reporting the error when using `end_offset=NUL... | diff --git a/tsl/test/expected/cagg_policy.out b/tsl/test/expected/cagg_policy.out
index aaa6d5b8621..d85f7f4a24c 100644
--- a/tsl/test/expected/cagg_policy.out
+++ b/tsl/test/expected/cagg_policy.out
@@ -1400,3 +1400,72 @@ SELECT timescaledb_experimental.add_policies('cagg');
f
(1 row)
+-- Issue #6902
+-- Fix tim... | [Bug]: Continuous aggregates with time_bucket hourly and timezone fail to update
### What type of bug is this?
Incorrect result, Unexpected error
### What subsystems and features are affected?
Continuous aggregate
### What happened?
I have a number of Continuous aggregates that uses hourly buckets
```
time_buck... | Thanks for the script, it reproduces for me. The error is related to the NULL value of the `end_offset`, so probably as a workaround you can set it to something far in the future but not null.
@fabriziomello should the end point calculation follow the same logic as `cagg_get_time_min`? Currently the null end point... | Thanks for the script, it reproduces for me. The error is related to the NULL value of the `end_offset`, so probably as a workaround you can set it to something far in the future but not null.
@fabriziomello should the end point calculation follow the same logic as `cagg_get_time_min`? Currently the null end point... | [
"https://github.com/timescale/timescaledb/commit/e39f398c62e9d3271ad15179f46aa4985ea54f4a"
] | 2025-08-28T23:08:57Z | https://github.com/timescale/timescaledb/tree/bbbebbfa978465bfdf6d0bac418ee4d269ad4da8 | [
"cmake --build build --parallel ; make -C build install"
] | [
"PATH=$PATH:/usr/lib/postgresql/17/bin pkill -u postgres postgres || true && rm -rf /tmp/pgdata && chown -R postgres:postgres /testbed/build && su - postgres -c \"PATH=$PATH:/usr/lib/postgresql/17/bin make -C /testbed/build installcheck\" 2>&1 | tee test-output.log"
] | [
"cat test-output.log"
] | def parser(log: str) -> dict[str, str]:
import re
results = {}
for line in log.splitlines():
m = re.match(r'(ok|not ok)\s+\d+\s+\+?\s*([^\s]+)', line)
if m:
status_word, test_name = m.groups()
status = "pass" if status_word == "ok" else "fail"
results[test... | [
"cagg_policy"
] | [
"compression_qualpushdown",
"chunk_column_stats",
"cagg_deprecated_bucket_ng",
"compress_default",
"cagg_refresh_using_trigger",
"cagg_policy_concurrent",
"compression_sorted_merge",
"modify_exclusion",
"ordered_append-17",
"compress_auto_sparse_index",
"vector_agg_groupagg",
"merge_append_par... | starryzhang/sweb.eval.x86_64.timescale_1776_timescaledb-8559 |
cilium/tetragon | 4069 | cilium__tetragon-4069 | [
"4056"
] | 67530967429bb42b665a7c7e69cdbc200247ccb4 | "diff --git a/api/v1/README.md b/api/v1/README.md\nindex 6bf750caea6..85596fcf70f 100644\n--- a/api/(...TRUNCATED) | "diff --git a/pkg/process/podinfo_test.go b/pkg/process/podinfo_test.go\nindex 46dba2c18b0..b06ec51f(...TRUNCATED) | "Include pod UID in the log\n### Is there an existing issue for this?\n\n- [x] I have searched the e(...TRUNCATED) | "As discussed in Slack, to implement this you can get inspiration from this https://github.com/ciliu(...TRUNCATED) | "As discussed in Slack, to implement this you can get inspiration from this https://github.com/ciliu(...TRUNCATED) | [
"https://github.com/cilium/tetragon/commit/3d566e06298cffff24f369932ac92e59f3049878"
] | 2025-08-29T05:43:42Z | https://github.com/cilium/tetragon/tree/67530967429bb42b665a7c7e69cdbc200247ccb4 | [
"export PATH=/usr/local/go/bin:$PATH ; make vendor ; make tetragon"
] | [
"go test ./... -v 2>&1 | tee test-output.log"
] | [
"cat test-output.log"
] | "def parser(log: str) -> dict[str, str]:\n import re\n results = {}\n pattern = re.compile((...TRUNCATED) | [
"TestK8sWatcher_GetPodInfo",
"TestProcessCache"
] | ["TestTracepointFieldParsing","FuzzProtojsonCompatibility/636add9a0930f826","FuzzProtojsonCompatibil(...TRUNCATED) | starryzhang/sweb.eval.x86_64.cilium_1776_tetragon-4069 |
quickjs-ng/quickjs | 1113 | quickjs-ng__quickjs-1113 | [
"1112"
] | 3d3b58d8815f0eef8122fad142a65f3bd21d273b | "diff --git a/quickjs.c b/quickjs.c\nindex 8b9a3b573..01eb10123 100644\n--- a/quickjs.c\n+++ b/quick(...TRUNCATED) | "diff --git a/tests/test_builtin.js b/tests/test_builtin.js\nindex 3c37f5930..2e86cb4a2 100644\n--- (...TRUNCATED) | "Cannot extend TypedArray\nThis should work:\n```js\nvar TypedArray = Object.getPrototypeOf(Uint8Arr(...TRUNCATED) | [
"https://github.com/quickjs-ng/quickjs/commit/c5e673abf837fced6d324174d653dae7a8584814"
] | 2025-06-24T20:23:51Z | https://github.com/quickjs-ng/quickjs/tree/3d3b58d8815f0eef8122fad142a65f3bd21d273b | [
"git submodule update --init --checkout --depth 1 ; make"
] | ["./build/run-test262 -vv -m -a -c test262.conf -c test262-fast.conf test262 2>&1 | tee test-output.(...TRUNCATED) | [
"cat test-output.log"
] | "def parser(log: str) -> dict[str, str]:\n import re\n results: dict[str, str] = {}\n # Mat(...TRUNCATED) | ["test262/test/language/expressions/class/dstr/async-gen-meth-dflt-ary-init-iter-get-err.js","test26(...TRUNCATED) | ["test262/test/built-ins/Array/fromAsync/this-constructor-with-unsettable-element-closes-sync-iterat(...TRUNCATED) | starryzhang/sweb.eval.x86_64.quickjs-ng_1776_quickjs-1113 | ||
fluent/fluent-bit | 10563 | fluent__fluent-bit-10563 | [
"10560"
] | c872957b57b2a8704e3b8cbc7f3994b430f96140 | "diff --git a/plugins/out_loki/loki.c b/plugins/out_loki/loki.c\nindex 2bdcaba3539..ee7a4313a89 1006(...TRUNCATED) | "diff --git a/tests/runtime/out_loki.c b/tests/runtime/out_loki.c\nindex a440fc0b692..e177b01e675 10(...TRUNCATED) | "Inconsistent `Remove_Keys` behavior when running Loki output with multiple workers / threads\n## Bu(...TRUNCATED) | Wow, that was fast! Thanks!
| ["https://github.com/fluent/fluent-bit/commit/f54cd8193712aed8eb64af38928ed76ef76e619f","https://git(...TRUNCATED) | 2025-07-09T04:16:04Z | https://github.com/fluent/fluent-bit/tree/c872957b57b2a8704e3b8cbc7f3994b430f96140 | ["cmake -S . -B build -G Ninja -DCMAKE_BUILD_TYPE=Release -DFLB_DEV=On -DFLB_TESTS_RUNTIME=On -DFLB_(...TRUNCATED) | [
"ctest --test-dir build --output-on-failure | tee test-output.log"
] | [
"cat test-output.log"
] | "def parser(log: str) -> dict[str, str]:\n import re\n results: dict[str, str] = {}\n # Mat(...TRUNCATED) | [
"flb-rt-out_stackdriver"
] | ["flb-rt-out_null","flb-rt-out_stdout","flb-rt-core_accept_timeout","flb-rt-custom_calyptia_input_te(...TRUNCATED) | starryzhang/sweb.eval.x86_64.fluent_1776_fluent-bit-10563 | |
php/php-src | 19478 | php__php-src-19478 | [
"19476"
] | 9c754baa99c795a65f40b94e4183150889895218 | "diff --git a/NEWS b/NEWS\nindex 77c1e32ff3a82..0180f5b3478a6 100644\n--- a/NEWS\n+++ b/NEWS\n@@ -5,(...TRUNCATED) | "diff --git a/Zend/tests/pipe_operator_reference_context.phpt b/Zend/tests/pipe_operator_reference_c(...TRUNCATED) | "pipe operator (`|>`) fails to correctly handle returning by reference\n### Description\n\nThere app(...TRUNCATED) | [
"https://github.com/php/php-src/commit/784755f5defdcaee7b7265bbf5c2d2c1ec3d2e42"
] | 2025-08-14T11:24:49Z | https://github.com/php/php-src/tree/9c754baa99c795a65f40b94e4183150889895218 | ["apt-get install -y pkg-config build-essential autoconf bison re2c libxml2-dev libsqlite3-dev ; ./c(...TRUNCATED) | [
"make test 2>&1 | tee test-output.log"
] | [
"cat test-output.log"
] | "def parser(log: str) -> dict[str, str]:\n import re\n results = {}\n # Match lines like: T(...TRUNCATED) | [
"Zend/tests/pipe_operator_reference_context.phpt"
] | ["Zend/tests/dynamic_prop_deprecation_002.phpt","ext/standard/tests/strings/utf8.phpt","tests/lang/b(...TRUNCATED) | starryzhang/sweb.eval.x86_64.php_1776_php-src-19478 | ||
varnishcache/varnish-cache | 4370 | varnishcache__varnish-cache-4370 | [
"4329"
] | c3b45a3140d340b2ed1a529b41253c2eb4ba70ae | "diff --git a/bin/varnishd/cache/cache_req_fsm.c b/bin/varnishd/cache/cache_req_fsm.c\nindex 8038102(...TRUNCATED) | "diff --git a/bin/varnishtest/tests/c00055.vtc b/bin/varnishtest/tests/c00055.vtc\nindex 0484219671.(...TRUNCATED) | "Bad chunked encoding should cause 400 error, not 503\n### Expected Behavior\n\nBad chunked encoding(...TRUNCATED) | bugwash agrees.
| bugwash agrees.
| ["https://github.com/varnishcache/varnish-cache/commit/dbdf05e0b7ec5cb7d4f7f8e991ebc670e011c257","ht(...TRUNCATED) | 2025-07-21T12:26:30Z | https://github.com/varnishcache/varnish-cache/tree/c3b45a3140d340b2ed1a529b41253c2eb4ba70ae | [
"./configure ; make -j$(nproc)"
] | [
"make check VERBOSE=1 2>&1 | tee test-output.log"
] | [
"cat test-output.log"
] | "def parser(log: str) -> dict[str, str]:\n import re\n results = {}\n for line in log.split(...TRUNCATED) | [
"tests/r02722.vtc",
"tests/f00001.vtc",
"tests/c00067.vtc",
"tests/f00016.vtc",
"tests/c00055.vtc"
] | ["vtest2/tests/a00002.vtc","vtest2/tests/a02016.vtc","tests/b00073.vtc","tests/r00502.vtc","tests/r0(...TRUNCATED) | starryzhang/sweb.eval.x86_64.varnishcache_1776_varnish-cache-4370 |
rsyslog/rsyslog | 6047 | rsyslog__rsyslog-6047 | [
"2424"
] | 4b3ce60b76362c16f0bece97bafc57fe4aaa3922 | "diff --git a/grammar/rainerscript.c b/grammar/rainerscript.c\nindex 82638305f4..b435b393f3 100644\n(...TRUNCATED) | "diff --git a/tests/rscript_random_warning.sh b/tests/rscript_random_warning.sh\nnew file mode 10075(...TRUNCATED) | "rainerscript: random() function does not provide user-visible warning when passed an unsupported va(...TRUNCATED) | [
"https://github.com/rsyslog/rsyslog/commit/2f62e6e76b961ae755d4103d60c010af42eedd7b"
] | 2025-08-28T10:26:55Z | https://github.com/rsyslog/rsyslog/tree/4b3ce60b76362c16f0bece97bafc57fe4aaa3922 | ["autoreconf -fvi ; ./configure --enable-imfile --enable-mysql --enable-usertools --enable-pgsql --e(...TRUNCATED) | ["./configure --enable-testbench --enable-imdiag --enable-omstdout && make -j$(nproc) && make check (...TRUNCATED) | [
"cat test-output.log"
] | "def parser(log: str) -> dict[str, str]:\n import re\n results = {}\n for line in log.split(...TRUNCATED) | [
"omfwd-lb-1target-retry-full_buf.sh"
] | ["rscript_substring.sh","rscript_set_modify.sh","imuxsock_logger_err.sh","func-substring-invld-start(...TRUNCATED) | starryzhang/sweb.eval.x86_64.rsyslog_1776_rsyslog-6047 | ||
aws/s2n-tls | 5481 | aws__s2n-tls-5481 | [
"5477"
] | 795f4dc65335b56dfb356d6eaac9429fbae5b75b | "diff --git a/tls/s2n_auth_selection.c b/tls/s2n_auth_selection.c\nindex 5678f4ae82a..6971f9495e3 10(...TRUNCATED) | "diff --git a/tests/integrationv2/common.py b/tests/integrationv2/common.py\nindex 5b526a27fbd..2667(...TRUNCATED) | "s2n-tls server reports signature scheme with RSA kex\n### Security issue notifications\n\nIf you di(...TRUNCATED) | ["https://github.com/aws/s2n-tls/commit/4f5133dc6c56030fd2b9691f75fd5e7f6259df09","https://github.co(...TRUNCATED) | 2025-08-25T16:58:39Z | https://github.com/aws/s2n-tls/tree/795f4dc65335b56dfb356d6eaac9429fbae5b75b | ["apt-get install -y libssl-dev ; cmake -S . -B build -G Ninja -DCMAKE_BUILD_TYPE=Release ; cmake --(...TRUNCATED) | [
"ctest --test-dir build --verbose 2>&1 | tee test-output.log"
] | [
"cat test-output.log"
] | "def parser(log: str) -> dict[str, str]:\n import re\n results: dict[str, str] = {}\n # Mat(...TRUNCATED) | [
"s2n_signature_algorithms_test",
"s2n_handshake_test"
] | ["s2n_client_signature_algorithms_extension_test","s2n_tls13_server_cert_test","s2n_client_session_t(...TRUNCATED) | starryzhang/sweb.eval.x86_64.aws_1776_s2n-tls-5481 |
End of preview. Expand in Data Studio
This is the multi - language version of SWE-bench-Live (repository_link ; paper_link).
We crawled GitHub issues as SWE tasks for coding agents to solve.
We used RepoLaunch (repository_link) to setup docker images as the problem-solving sandboxes for each instance.
Each split is task set of one language.
Each split is filtered by prompting LLM -- a task instance is discarded if LLM cannot infer test patch from problem statement or problem statement contains solution (in NL or code).
Dataset Schema
| Field | Type |
|---|---|
| instance_id | str |
| repo | str |
| pull_number | str |
| issue_numbers | list[str] |
| base_commit | str |
| problem_statement | str |
| patch | str |
| test_patch | str |
| hints_text | Optional[str] from the discussions of the issue |
| all_hints_text | Optional[str] |
| commit_urls | list[str] |
| created_at | str in ISO datetime format |
| commit_url | str |
| rebuild_cmds | list[str] |
| test_cmds | list[str] |
| log_parser | str |
| FAIL_TO_PASS | list[str] |
| PASS_TO_PASS | list[str] |
| docker_image | str |
Evaluation Protocol
SWE-bench-Live evaluation strictly follows the original
SWE-benchprotocol:
- During a rollout, the agent may access only the
problem_statementfield of the Hugging Face dataset and the docker image of the task instance. It must not access any other fields, such ashint,FAIL_TO_PASS, ortest_patch. Thetest_patchmust not be applied to the repository before or during the rollout. The agent must perform a single rollout based solely on theproblem_statementon the docker container started from the image of the task instance.- Prompts, skills, and workflow instructions provided to the agent must not contain solutions specific to any task instance. They may contain only general instructions for the entire benchmark or, at most, for a specific repository. The agent must not use results from the ground-truth evaluation script to refine its solution.
Compliant prompts should follow the SWE-agent prompt and the OpenHands prompt, which contain only the problem statement and general workflow instructions.
When submitting results to our submissions repository, you must include your agent's raw rollout trajectories so that the maintainers can verify compliance with the SWE-bench protocol. A trajectory consists of the complete sequence of inputs to and outputs from your agent across all rollout rounds for a given task instance, including the initial prompt provided to the agent. Please follow this SWE-agent compliant trajectory example when submitting your result. If your organization's policies prohibit sharing the complete set of trajectories, you must provide at least some representative samples for verification. There is a checklist when submitting a PR to help you check whether you meet the protocol requirements again.
- Downloads last month
- 24,104