tuieval 0.2.0.dev112__py3-none-any.whl → 0.2.0.dev118__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
tuieval/_version.py CHANGED
@@ -18,7 +18,7 @@ version_tuple: tuple[int | str, ...]
18
18
  commit_id: str | None
19
19
  __commit_id__: str | None
20
20
 
21
- __version__ = version = '0.2.0.dev112'
22
- __version_tuple__ = version_tuple = (0, 2, 0, 'dev112')
21
+ __version__ = version = '0.2.0.dev118'
22
+ __version_tuple__ = version_tuple = (0, 2, 0, 'dev118')
23
23
 
24
24
  __commit_id__ = commit_id = None
tuieval/client.py CHANGED
@@ -118,6 +118,7 @@ def stream_chat(base_url, body, on_delta=None, timeout=1800, stream=None, header
118
118
  out = {"answer": "", "reasoning": "", "tool_calls": [], "finish": None, "usage": {}, "timings": {},
119
119
  "ttft_s": None, "total_s": None, "error": None}
120
120
  reasoning, answer, tools = [], [], {}
121
+ done = False # [DONE] seen
121
122
  start = last = time.time()
122
123
  try:
123
124
  conn.request("POST", prefix + "/v1/chat/completions", body=json.dumps(body),
@@ -142,6 +143,7 @@ def stream_chat(base_url, body, on_delta=None, timeout=1800, stream=None, header
142
143
  continue
143
144
  payload = line[5:].strip()
144
145
  if payload == "[DONE]":
146
+ done = True
145
147
  break
146
148
  chunk = json.loads(payload)
147
149
  if chunk.get("error"): # OpenRouter reports mid-stream failures this way
@@ -173,6 +175,9 @@ def stream_chat(base_url, body, on_delta=None, timeout=1800, stream=None, header
173
175
  if text:
174
176
  answer.append(text)
175
177
  on_delta and on_delta("answer", text)
178
+ if not out["error"] and not done and out["finish"] is None:
179
+ # the stream just ended: a server that crashed or dropped the connection, not the model's answer
180
+ out["error"] = "connection error: the stream ended before the server finished (no finish reason)"
176
181
  except (OSError, http.client.HTTPException, ValueError) as e:
177
182
  if stream.cancelled:
178
183
  raise Cancelled() from None
tuieval/dates.py ADDED
@@ -0,0 +1,164 @@
1
+ """When each answer was given, and when each test was added to its pack.
2
+
3
+ Answers record `answered_at` (local time with its UTC offset). Older answers have none, but their
4
+ log file name starts with the time the answer finished (logs/live/20261014_143741_…), so they get it
5
+ from that.
6
+
7
+ A test's date is its `added` field when it has one (tests.yaml: `added: 2026-10-14`). Otherwise
8
+ tuieval notes the day it first sees the test, in test-dates.json at the workspace root. Tests that were
9
+ already there when that file was started get the day of their first answer, or the day the file was
10
+ started, as "on or before" dates: the day they were really written isn't known.
11
+ """
12
+ import datetime
13
+ import glob
14
+ import json
15
+ import os
16
+ import re
17
+
18
+ FILE = "test-dates.json"
19
+ _LOG_STAMP = re.compile(r"(\d{8}_\d{6})_")
20
+ _HISTORY = re.compile(r"-\d{8}-\d{6}(\.partial\.jsonl|\.json)$")
21
+
22
+
23
+ def now():
24
+ return datetime.datetime.now().astimezone().isoformat(timespec="seconds")
25
+
26
+
27
+ def answered_at(rec):
28
+ """When an answer finished (an aware datetime), or None if its result doesn't say."""
29
+ if rec.get("answered_at"):
30
+ try:
31
+ return datetime.datetime.fromisoformat(rec["answered_at"])
32
+ except ValueError:
33
+ return None
34
+ m = _LOG_STAMP.match(rec.get("log") or "")
35
+ if m:
36
+ try:
37
+ return datetime.datetime.strptime(m.group(1), "%Y%m%d_%H%M%S").astimezone()
38
+ except ValueError:
39
+ return None
40
+ return None
41
+
42
+
43
+ def day(dt):
44
+ return dt.strftime("%Y-%m-%d") if dt else ""
45
+
46
+
47
+ def minute(dt):
48
+ return dt.strftime("%Y-%m-%d %H:%M") if dt else ""
49
+
50
+
51
+ def span(dts):
52
+ """'2026-10-14', or '2026-10-01 – 10-07' for answers given over several days."""
53
+ dts = [d for d in dts if d]
54
+ if not dts:
55
+ return ""
56
+ first, last = day(min(dts)), day(max(dts))
57
+ if first == last:
58
+ return first
59
+ return f"{first} – {last[5:] if first[:4] == last[:4] else last}"
60
+
61
+
62
+ def _explicit(test):
63
+ v = test.get("added")
64
+ if isinstance(v, datetime.datetime):
65
+ return v.date().isoformat()
66
+ if isinstance(v, datetime.date):
67
+ return v.isoformat()
68
+ if isinstance(v, str):
69
+ try:
70
+ return datetime.date.fromisoformat(v.strip()).isoformat()
71
+ except ValueError:
72
+ return None
73
+ return None
74
+
75
+
76
+ def first_answers(results_dir):
77
+ """{(pack, test id): the earliest day it was answered} over every result, history and archive."""
78
+ first = {}
79
+
80
+ def note(pack, rec):
81
+ dt = answered_at(rec)
82
+ if pack and dt and rec.get("test"):
83
+ key, d = (pack, rec["test"]), day(dt)
84
+ if key not in first or d < first[key]:
85
+ first[key] = d
86
+ for path in glob.glob(os.path.join(glob.escape(results_dir), "**", "*.json*"), recursive=True):
87
+ base = os.path.basename(path)
88
+ try:
89
+ if base.endswith(".partial.jsonl"):
90
+ pack = _HISTORY.sub("", base) if _HISTORY.search(base) else base[:-len(".json.partial.jsonl")]
91
+ with open(path) as f:
92
+ for line in f:
93
+ rec = json.loads(line)
94
+ if "test" in rec:
95
+ note(pack, rec)
96
+ elif base.endswith(".json"):
97
+ with open(path) as f:
98
+ data = json.load(f)
99
+ if isinstance(data, dict) and data.get("format") == "evals/1":
100
+ for rec in data.get("results", []):
101
+ note(data["pack"]["name"], rec)
102
+ except (OSError, ValueError, KeyError, TypeError):
103
+ continue
104
+ return first
105
+
106
+
107
+ class AddedDates:
108
+ """Each test's added date: {(pack, test id): (YYYY-MM-DD, exact)}. Not exact = on or before."""
109
+
110
+ def __init__(self, root, packs, results_dir):
111
+ self.path = os.path.join(root, FILE)
112
+ try:
113
+ with open(self.path) as f:
114
+ saved = json.load(f).get("tests", {})
115
+ started = True
116
+ except (OSError, ValueError, AttributeError):
117
+ saved, started = {}, False
118
+ self.dates = {}
119
+ missing = []
120
+ for name, pack in packs.items():
121
+ for t in pack.tests:
122
+ key = (name, t["id"])
123
+ given = _explicit(t)
124
+ if given:
125
+ self.dates[key] = (given, True)
126
+ elif f"{name}/{t['id']}" in saved:
127
+ s = saved[f"{name}/{t['id']}"]
128
+ self.dates[key] = (s["added"], s.get("how") == "seen")
129
+ else:
130
+ missing.append(key)
131
+ if not missing:
132
+ return
133
+ today = datetime.date.today().isoformat()
134
+ first = first_answers(results_dir)
135
+ for key in missing:
136
+ if key in first: # answered before tuieval noted it: on or before its first answer
137
+ saved[f"{key[0]}/{key[1]}"] = {"added": first[key], "how": "first answer"}
138
+ elif started: # new since the file was started
139
+ saved[f"{key[0]}/{key[1]}"] = {"added": today, "how": "seen"}
140
+ else: # already there when the file was started
141
+ saved[f"{key[0]}/{key[1]}"] = {"added": today, "how": "already there"}
142
+ s = saved[f"{key[0]}/{key[1]}"]
143
+ self.dates[key] = (s["added"], s["how"] == "seen")
144
+ try:
145
+ tmp = self.path + ".tmp"
146
+ with open(tmp, "w") as f:
147
+ json.dump({"about": "The day tuieval first saw each test (how: seen), or for tests it found "
148
+ "already there, a day on or before it. A test's own `added` field wins.",
149
+ "tests": dict(sorted(saved.items()))}, f, indent=1)
150
+ f.write("\n")
151
+ os.replace(tmp, self.path)
152
+ except OSError:
153
+ pass # a read-only workspace: the dates still show, they just aren't kept
154
+
155
+ def get(self, pack, test_id):
156
+ """(YYYY-MM-DD, exact) or None."""
157
+ return self.dates.get((pack, test_id))
158
+
159
+ def text(self, pack, test_id):
160
+ """'2026-10-14', '≤ 2026-10-01' (on or before), or ''."""
161
+ got = self.get(pack, test_id)
162
+ if not got:
163
+ return ""
164
+ return got[0] if got[1] else f"≤ {got[0]}"
tuieval/engine.py CHANGED
@@ -51,6 +51,7 @@ import urllib.parse
51
51
  import urllib.request
52
52
 
53
53
  from . import client
54
+ from . import dates
54
55
  from . import graders
55
56
  from . import machines
56
57
  from . import packs as packs_mod
@@ -714,6 +715,7 @@ class Engine:
714
715
  self.cfg = load_config(self.models_path)
715
716
  self.pack_errors = []
716
717
  self.packs = packs_mod.load_packs(self.packs_dir, self.pack_errors)
718
+ self.test_dates = dates.AddedDates(root, self.packs, self.results_dir)
717
719
  self._cancel = threading.Event()
718
720
  self._main_lane = Lane()
719
721
  self._here = threading.local() # .lane: the lane this thread runs
@@ -766,6 +768,7 @@ class Engine:
766
768
  self._add_remote_models()
767
769
  self.pack_errors = []
768
770
  self.packs = packs_mod.load_packs(self.packs_dir, self.pack_errors)
771
+ self.test_dates = dates.AddedDates(self.root, self.packs, self.results_dir)
769
772
 
770
773
  def emit(self, kind, **data):
771
774
  try:
@@ -2348,6 +2351,7 @@ class Engine:
2348
2351
  "severity": g.get("severity"), "answer": res["answer"], "tool_calls": res["tool_calls"],
2349
2352
  "finish": res["finish"], "reasoning_chars": len(res["reasoning"]),
2350
2353
  "log": log_name, # logs/live/<log>: the reasoning, which the result doesn't keep
2354
+ "answered_at": dates.now(),
2351
2355
  **({"checks": g["checks"]} if "checks" in g else {}),
2352
2356
  "machine": self.machine().id,
2353
2357
  **({"quantization": ep.get("quantization")} if ep else {}),
tuieval/packs.py CHANGED
@@ -132,7 +132,7 @@ def _read_tests(path):
132
132
  # Fields that never reach the model or the grader's decision: editing them doesn't make
133
133
  # results outdated.
134
134
  NOT_FINGERPRINTED = {"reference", "wrong", "description", "category", "critical", "critical_trial", "group",
135
- "captured_from", "captured_answer", "difficulty"}
135
+ "captured_from", "captured_answer", "difficulty", "added"}
136
136
  DIFFICULTIES = ("easy", "medium", "hard")
137
137
 
138
138
 
tuieval/tui.py CHANGED
@@ -45,6 +45,7 @@ from rich.cells import cell_len
45
45
  from rich.text import Text
46
46
 
47
47
  from . import compare
48
+ from . import dates
48
49
  from . import engine
49
50
  from . import verdict
50
51
 
@@ -618,11 +619,14 @@ class AnswerScreen(ModalScreen):
618
619
  f"on {rec['machine']}" if rec.get("machine") else "",
619
620
  f"[yellow]ran alongside {', '.join(rec['ran_alongside'])}[/yellow]" if rec.get("ran_alongside") else "",
620
621
  f"seed {rec['seed']}" if "seed" in rec else ""]
622
+ when = dates.minute(dates.answered_at(rec))
623
+ added = e.test_dates.text(entry["pack"], rec["test"])
621
624
  self.query_one("#answer-head", Static).update(
622
625
  f"{mark} {diff_badge(rec.get('difficulty', 'unrated'))} [b]{rich_escape(rec.get('description', rec['test']))}[/b]"
623
626
  f"{partial}{pos}\n"
624
627
  f"[b]{entry['model']}[/b]{quant} · {pack.label if pack else entry['pack']} · {rec['test']} · "
625
- f"repeat {rec.get('repeat', 0) + 1}\n"
628
+ f"repeat {rec.get('repeat', 0) + 1}" + (f" · answered {when}" if when else "")
629
+ + (f" · test added {added}" if added else "") + "\n"
626
630
  + " · ".join(f for f in facts if f))
627
631
  grading = [f"[b]Grading:[/b] {rich_escape(rec.get('reason', ''))}"]
628
632
  grading += [" " + line for line in check_lines(rec, test)]
@@ -1986,7 +1990,9 @@ class PickTestsScreen(ModalScreen):
1986
1990
  mark = "[b green]\\[✓][/]" if on else "[dim]\\[ ][/dim]"
1987
1991
  tid = f"[b]{rich_escape(t['id'])}[/b]" if on else rich_escape(t["id"])
1988
1992
  note = test_note(t)
1989
- return f"{mark} {diff_badge(t.get('difficulty', 'unrated'))} {tid}" + (f" [dim]{rich_escape(note)}[/dim]" if note else "")
1993
+ added = self.app.engine.test_dates.text(self.pack.name, t["id"])
1994
+ return f"{mark} {diff_badge(t.get('difficulty', 'unrated'))} {tid}" + (f" [dim]{rich_escape(note)}[/dim]" if note else "") \
1995
+ + (f" [dim]added {added}[/dim]" if added else "")
1990
1996
 
1991
1997
  def fill(self, text):
1992
1998
  ol = self.query_one("#pick-tests", OptionList)
@@ -2998,14 +3004,15 @@ class ResultsScreen(Screen):
2998
3004
  """Every wrong answer, of the models picked (Models: shared with the other tabs)."""
2999
3005
  f = self.query_one("#failures", DataTable)
3000
3006
  f.clear(columns=True)
3001
- f.add_columns("Model", "Level", "Test", "Reason")
3007
+ f.add_columns("Model", "Level", "Test", "Answered", "Reason")
3002
3008
  every = {r["model"] for r in self.rows}
3003
3009
  chosen = {m for m in self.app.pq_models if m in every}
3004
3010
  rows = [r for r in self.rows if not chosen or r["model"] in chosen]
3005
3011
  # compare.failures lists the failed rows in order, so they line up with fail_rows
3006
3012
  self.fail_rows = [r for r in rows if not r["ok"]]
3007
3013
  for i, (model, test, reason, level) in enumerate(compare.failures(rows)):
3008
- f.add_row(model, diff_badge(level), rich_escape(test), rich_escape(reason[:130]), key=str(i))
3014
+ when = dates.day(dates.answered_at(self.fail_rows[i].get("raw") or {}))
3015
+ f.add_row(model, diff_badge(level), rich_escape(test), when, rich_escape(reason[:130]), key=str(i))
3009
3016
  self.query_one("#fail-models", Button).label = f"Models: {len(chosen)} picked" if chosen else "Models: all"
3010
3017
  n = len(self.fail_rows)
3011
3018
  self.query_one("#fail-count", Static).update(
@@ -3283,7 +3290,7 @@ class ResultsScreen(Screen):
3283
3290
  here = self.app.engine.machine().id
3284
3291
  table = self.query_one("#per-question", DataTable)
3285
3292
  table.clear(columns=True)
3286
- table.add_columns("Test", "Pack", "Level", "×tokens", *models)
3293
+ table.add_columns("Test", "Pack", "Level", "Added", "×tokens", *models)
3287
3294
  self.pq_tests, self.pq_models = {}, models
3288
3295
  for t in tests:
3289
3296
  cells = []
@@ -3295,7 +3302,9 @@ class ResultsScreen(Screen):
3295
3302
  spread = t["spread"]["tokens"]
3296
3303
  sp = f"[yellow]{spread:.1f}×[/yellow]" if spread >= 3 else (f"{spread:.1f}×" if len(t["cells"]) > 1 else "")
3297
3304
  desc = t["cells"][next(iter(t["cells"]))]["rows"][0].get("raw", {}).get("description") or t["name"]
3298
- table.add_row(rich_escape(desc[:48]), t["suite"], diff_badge(t["difficulty"]), sp, *cells, key=t["test"])
3305
+ added = self.app.engine.test_dates.get(t["suite"], t["test"].split(": ", 1)[-1])
3306
+ added = sort_cell(added[0] if added[1] else f"≤ {added[0]}", added[0]) if added else ""
3307
+ table.add_row(rich_escape(desc[:48]), t["suite"], diff_badge(t["difficulty"]), added, sp, *cells, key=t["test"])
3299
3308
  self.pq_tests[t["test"]] = t
3300
3309
  self.sort_table(table)
3301
3310
  self.sort_table(self.query_one("#pq-summary", DataTable))
@@ -3434,9 +3443,12 @@ class ResultsScreen(Screen):
3434
3443
  "[dim]No verdicts yet: run something from Setup.[/dim]")
3435
3444
  groups = sorted({g for t in table.values() for g in t})
3436
3445
  matrix.add_columns("Model", *groups)
3437
- detail.add_columns("Model", "Use case", "Pack", "Verdict", "Answers right", "Why")
3446
+ detail.add_columns("Model", "Use case", "Pack", "Verdict", "Answers right", "Answered", "Why")
3438
3447
  right = collections.Counter((r["model"], r["suite"]) for r in self.rows if r["ok"])
3439
3448
  answered = collections.Counter((r["model"], r["suite"]) for r in self.rows)
3449
+ when = collections.defaultdict(list)
3450
+ for r in self.rows:
3451
+ when[(r["model"], r["suite"])].append(dates.answered_at(r.get("raw") or {}))
3440
3452
  lat = verdict.latency_table(self.app.engine, None, table)
3441
3453
  machine_ids = [mc.id for mc in self.app.engine.machines()]
3442
3454
  lt = self.query_one("#latency", DataTable)
@@ -3455,7 +3467,8 @@ class ResultsScreen(Screen):
3455
3467
  if pv.status != "NO DATA":
3456
3468
  n = answered[(label, name)]
3457
3469
  detail.add_row(label, g, name, f"[{style[pv.status]}]{pv.status}[/]",
3458
- f"{right[(label, name)]}/{n}" if n else "-", "; ".join(pv.reasons)[:160])
3470
+ f"{right[(label, name)]}/{n}" if n else "-", dates.span(when[(label, name)]),
3471
+ "; ".join(pv.reasons)[:160])
3459
3472
 
3460
3473
 
3461
3474
  # ------------------------------------------------------------------ app
tuieval/tune.py CHANGED
@@ -169,7 +169,37 @@ def _body(m, sampling, msgs, max_tokens=GEN_TOKENS, request=None):
169
169
  **(request or {})}
170
170
 
171
171
 
172
+ def _rejects_ignore_eos(res):
173
+ """True if the server refused the request because it doesn't accept the ignore_eos field."""
174
+ err = res.get("error") or ""
175
+ return err.startswith("server returned HTTP 4") and "ignore_eos" in err
176
+
177
+
172
178
  def _request(eng, base_url, body, name, emit, limit):
179
+ """One streamed request with live progress and a wall-clock limit. A server that refuses
180
+ ignore_eos is asked again without it, and later requests leave it out (see no_ignore_eos)."""
181
+ key = (base_url, body.get("model"))
182
+ rejected = no_ignore_eos(eng)
183
+ if key in rejected:
184
+ body = {k: v for k, v in body.items() if k != "ignore_eos"}
185
+ res = _stream_request(eng, base_url, body, name, emit, limit)
186
+ if body.get("ignore_eos") and _rejects_ignore_eos(res):
187
+ rejected.add(key)
188
+ emit("tune_step", message=" the server doesn't accept ignore_eos: answers may end early "
189
+ "(speeds are per token, so still comparable)")
190
+ res = _stream_request(eng, base_url, {k: v for k, v in body.items() if k != "ignore_eos"},
191
+ name, emit, limit)
192
+ return res
193
+
194
+
195
+ def no_ignore_eos(eng):
196
+ """(base_url, served model) pairs whose server refused ignore_eos in this tuning session."""
197
+ if not hasattr(eng, "_tune_no_ignore_eos"):
198
+ eng._tune_no_ignore_eos = set()
199
+ return eng._tune_no_ignore_eos
200
+
201
+
202
+ def _stream_request(eng, base_url, body, name, emit, limit):
173
203
  """One streamed request with live progress (every PROGRESS_EVERY_S: waiting for the first
174
204
  token, or ~tokens so far and tok/s) and a wall-clock limit. Returns client.stream_chat's result."""
175
205
  st, t0 = client.Stream(), time.time()
@@ -681,6 +711,9 @@ def tune(eng, label, emit=lambda *a, **k: None, max_starts=16, min_gain=0.03, us
681
711
  if warm_info:
682
712
  notes.insert(0, f"warm start from {warm_info['from']}: inherited "
683
713
  f"{', '.join(warm_info['inherited']) or 'nothing'}")
714
+ if any(model == m["served_name"] for _, model in no_ignore_eos(eng)):
715
+ notes.insert(0, f"the server doesn't accept ignore_eos, so workload answers could end before "
716
+ f"{gen_tokens} tokens (speeds are per token)")
684
717
  if best_r.facts.get("extra_swap_mb") and best_r is not base:
685
718
  warnings.append(f"the chosen flags use more memory than the defaults: macOS swapped "
686
719
  f"~{best_r.facts['extra_swap_mb']} MB more of other apps' memory; close apps during runs, "
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.5
2
2
  Name: tuieval
3
- Version: 0.2.0.dev112
3
+ Version: 0.2.0.dev118
4
4
  Summary: Evaluate local and frontier LLMs on your own questions: accuracy, speed, tokens and PASS/FAIL verdicts, in the terminal.
5
5
  Project-URL: Homepage, https://github.com/ashe-wb/tuieval
6
6
  Project-URL: Issues, https://github.com/ashe-wb/tuieval/issues
@@ -1,26 +1,27 @@
1
1
  tuieval/__init__.py,sha256=myVXVX7htbwezW6k3x_FuKgn1RSZ2WcTaDEMByGn6Wg,312
2
2
  tuieval/__main__.py,sha256=E6Gls0DNz8GQK2K-kOUIx8cYhgANW_CH54VKrfCfs14,52
3
- tuieval/_version.py,sha256=W7ovaJXOHI4LeAQAjKdSNMOHKKyCHZRZ3K5cv33MJLk,537
3
+ tuieval/_version.py,sha256=Ghcc-_tDTmxerkJyex26IeC491wfUqj0PD6Wec9ZKMo,537
4
4
  tuieval/builtin.toml,sha256=Nwu7O0qFzcrDAxXO_VZtDPvjVwxZ29OtYOW0BPihZGs,4410
5
5
  tuieval/cli.py,sha256=mr0o_ud6NctZeU4WIrh17x_uzEK0vTeCynGP0wrCeBY,4812
6
- tuieval/client.py,sha256=wtACBVsngtWopWWhIc4bB6mEUEi31dvvDePN-PjmXbM,11122
6
+ tuieval/client.py,sha256=WhhIQwkKn1kVVV3DZqUR25INLuGFzSotjouPwmzJxQg,11489
7
7
  tuieval/compare.py,sha256=UjZcWsbPep7EG4XRKWaEthUT0l0iN8rKIs1135dbt6c,32139
8
+ tuieval/dates.py,sha256=CVazlwNjCRhD__D12zXLdeqF2ySt15X3xLUQeE-o6PY,6293
8
9
  tuieval/doctor.py,sha256=j2MdDORstsAGPnRmpsnxau1_fR6sTc_oEFGM5zugTJU,7246
9
- tuieval/engine.py,sha256=J_r60eRHkJYWFXkK5C0soYzeDkJqep0SS14GixC7NOE,127431
10
+ tuieval/engine.py,sha256=gO3SGyvF43Zu8fAlTSIzrQ7kwFNJmWgnATm4QmAOr9o,127665
10
11
  tuieval/export.py,sha256=dbHQgL_ubWtIPM1sLzAew_gETnrDJyd0zgqVA5vkpXM,9657
11
12
  tuieval/insights.py,sha256=m6HZk1VbX-avHr3xyT8eVJ5OLjKa_EKTVBxsu04X_HQ,28947
12
13
  tuieval/logs.py,sha256=01gZK6tf53dIThaIrs1bEsz9eqKYVlcxpyaLnSPTVxo,8127
13
14
  tuieval/machines.py,sha256=t9tu6CZeXZON4EtOVnEwP4VtaynpVLf9Cz8TNjf7jtY,20124
14
15
  tuieval/onboard.py,sha256=IYETVDWCCo9vF0Dtj4_rr9VanYeCky8PwP3-2g5A_l0,6215
15
- tuieval/packs.py,sha256=ZgaZ1m2ePYvDjmddSc3nafnMEn7kTq9WcDpCaJSZleQ,12026
16
+ tuieval/packs.py,sha256=cJZxCllaLqnpnqpcTqzFVl8BXnsL3852ZkFUa7Ki1jU,12035
16
17
  tuieval/profiles.py,sha256=soeLxjdB-l-OxSY8jZ4yjwfdSnsBV-QICcwKxRJS7nQ,3663
17
18
  tuieval/pta.py,sha256=nU0HLl28CoRlmU4Hw3rqy6i03e7bG0Eto_gShS_eqUM,13552
18
19
  tuieval/remove.py,sha256=Vwh43nm9lVb9U9aN-h6CMYbkTwdR3UWfZObwweF72n0,5157
19
20
  tuieval/run_evals.py,sha256=DN6FhC8Zj08N5pwlrCmlkbkTxA2vmT-G5sWJnp-S-AI,50114
20
21
  tuieval/scaffold.py,sha256=YBOqniJpf_GtKKAFF3_OHxxMGbmTWdFjcClGcOnrrKE,12130
21
22
  tuieval/selftest.py,sha256=nExkrOidgymlmvkxImEBfrQG5iTUN5WuRMIKtqusCCs,9835
22
- tuieval/tui.py,sha256=UFVL89F5AkoN7KUMoDfFI7cCZChMpoutwceJtz_4IrE,195167
23
- tuieval/tune.py,sha256=azxxthgJeag8JyoQUgaRxq6oI1XvEgzb9vkszjJZH7w,39645
23
+ tuieval/tui.py,sha256=_PjBy5KMKD0RiWrZ7Orr6Ge1c6wP8SkW4MIm-fqyrVw,196105
24
+ tuieval/tune.py,sha256=izz1Qs-d9PrWAqEZpyrLxfrlhlaouif9uFQSeJveL0k,41294
24
25
  tuieval/verdict.py,sha256=3SM8qrVdWNIFDENagaR5ZPBYEnjsBtkYB9WS5TDVssU,20517
25
26
  tuieval/watch_proxy.py,sha256=bnDO1i6n6yUs1hODoUWcMppJMGPavoQqs6aNYrIxpT8,11416
26
27
  tuieval/workspace.py,sha256=V_HhkaN1odr0OpjNsPYradRWLpzDldM7SwpvxohBL4U,895
@@ -48,8 +49,8 @@ tuieval/templates/packs/tool_call/pack.toml,sha256=n4bx0lGJHSmW41uVqhfIXJ2BWZpmT
48
49
  tuieval/templates/packs/tool_call/system.txt,sha256=mKxu-lB-0YjLHXdgeFJa8x2C180OvcNz5cHv2N5Bg3U,254
49
50
  tuieval/templates/packs/tool_call/tests.yaml,sha256=Nv7c4syrzJ9cU8EBKSK_1ZSFx9V02aCjW4-2TBCPT7s,1859
50
51
  tuieval/templates/packs/tool_call/tools.yaml,sha256=HgxCkh0On2VwbHSsjAGM8tNCK7ZtKLGrs7lMCn3Fprs,1048
51
- tuieval-0.2.0.dev112.dist-info/METADATA,sha256=toA2Ug8fgr_Fd--c7Qa4Lp9erPs35Ur-Cl1baVSDP2E,14798
52
- tuieval-0.2.0.dev112.dist-info/WHEEL,sha256=W3fkpkm7-wf9vBI5Z-7s0eWkeM-spu78I8Neb98DeEg,87
53
- tuieval-0.2.0.dev112.dist-info/entry_points.txt,sha256=7eIhi1GpFxBP2OLJV02Nn8LwsNuO512eKOGt1IBbj04,45
54
- tuieval-0.2.0.dev112.dist-info/licenses/LICENSE,sha256=7wfRgGtBGH1aXhZkIn47oUznqKC18kEPMvUVgol7Ffc,1077
55
- tuieval-0.2.0.dev112.dist-info/RECORD,,
52
+ tuieval-0.2.0.dev118.dist-info/METADATA,sha256=1qcgO4VMEhAGyeWxRHpGWKwZAL7yo_eFbxCe8cFrTvM,14798
53
+ tuieval-0.2.0.dev118.dist-info/WHEEL,sha256=W3fkpkm7-wf9vBI5Z-7s0eWkeM-spu78I8Neb98DeEg,87
54
+ tuieval-0.2.0.dev118.dist-info/entry_points.txt,sha256=7eIhi1GpFxBP2OLJV02Nn8LwsNuO512eKOGt1IBbj04,45
55
+ tuieval-0.2.0.dev118.dist-info/licenses/LICENSE,sha256=7wfRgGtBGH1aXhZkIn47oUznqKC18kEPMvUVgol7Ffc,1077
56
+ tuieval-0.2.0.dev118.dist-info/RECORD,,