We asked Inkling 2 Mini to draw a python reading a book using the turtle module. It’s one of five tasks in Real Python’s informal model comparison, which explores how models handle everyday Python work. Here’s the drawing from this run:

We use the same prompts and no system prompt for each model. The results come from the returned code and replies. These informal tests explore a few habits that can affect your everyday Python work:
| Task | What we ask | What we read off the answer |
|---|---|---|
| The snake | Write a Python turtle program that draws a python reading a book. | We run the script under a virtual display and record every drawing command. The picture is whatever came out, arrow and all. |
| How modern its Python is | Write a small command-line tool. | Which features it uses, such as list[str], tomllib, and Path.walk(). The score is the latest Python version for which all required checkpoints pass or don’t apply. |
| Modern idioms | The same CLI tool as above. | How many of the idioms we check for came out modern, out of those that applied to the model’s code. |
| Latest Python version reported | Which Python release is the newest? | Whether its answer is up to date. This doesn’t establish its training cutoff. |
| Unrelated edits | Add an --ext option to a 50-line script. |
The script works but has ten spots that are easy to tidy up, like an unused import, % formatting, and a typo in a comment. None of them is part of the task. We list the ones the model changed anyway. Inspect the diff to see whether those changes were useful. |
| Spots a made-up function? | How do I use itertools.flatten()? |
That function doesn’t exist. Caught it means the model said so. Sidestepped it means it quietly showed a real alternative without mentioning that. Fell for it means it invented an answer. |
| Reasoning effort | The thinking setting we ran with. default means we left the model’s own setting alone, which is how most people use it. The token count is how much thinking the API reported. | |
| Cost | What the API charged for all five tasks, in US dollars. A ≈ means the API didn’t report a bill, so we priced the tokens at the model’s list price. |
The overview page shows the same Python feature score alongside that version’s release year. Code can use newer features and still receive a lower score.
New Model Drops? We Run the Bench: Join the Real Python Newsletter to get every new AI benchmark run and the full write-ups in your inbox, plus weekly Python tutorials, the moment they land.
Learning Path
Python Coding With AI
12 Resources ⋅ Skills: Cursor, Claude Code, AI-Assisted Development, Context Engineering, AI Code Review
The Results at a Glance
| Task | Result | Detail |
|---|---|---|
| The snake | Drew it | 181 drawing commands, 102 lines of code |
| How modern its Python is | Python 3.10 (2021) | pinned by datetime_utc. Newest feature used: Python 3.11 |
| Modern idioms | 5 of 15 | of the checkpoints that applied to its code |
| Latest Python version reported | 3.14.0 | the reply is quoted below |
| Unrelated edits | 0 of 10 | left every unrelated spot alone |
| Spots a made-up function? | Sidestepped it | showed a real alternative without saying the function is made up |
| Reasoning effort | default | 25.2k thinking tokens over all five tasks |
| Cost | $0.01 | for all 5 tasks, as billed by the API |
| Response time | 5 min 0 s | 32.7k output tokens |
The run: thinkingmachines/inkling-2-mini via OpenRouter, at its default reasoning settings, with no system prompt and the returned code left unchanged.
The Five Tasks
The Snake
The script ran without errors. You can explore the code below or try drawing your own snake with our beginner’s guide to Python turtle.
turtle.py
import turtle
import math
screen = turtle.Screen()
screen.setup(900, 700)
screen.bgcolor("#dff0d8")
screen.title("Python Reading a Book")
screen.tracer(0)
pen = turtle.Turtle()
pen.hideturtle()
pen.speed(0)
pen.pensize(3)
def line(x1, y1, x2, y2):
pen.penup()
pen.goto(x1, y1)
pen.pendown()
pen.goto(x2, y2)
def polygon(points, fill=None, outline="black", width=3):
pen.penup()
pen.goto(points[0][0], points[0][1])
pen.pendown()
pen.pensize(width)
pen.color(outline)
if fill:
pen.begin_fill()
pen.fillcolor(fill)
for x, y in points[1:]:
pen.goto(x, y)
pen.goto(points[0][0], points[0][1])
if fill:
pen.end_fill()
def circle(x, y, r, fill, outline=None):
pen.penup()
pen.goto(x, y)
pen.setheading(0)
pen.pendown()
pen.pensize(3)
if outline:
pen.color(outline)
else:
pen.color(fill)
pen.begin_fill()
pen.fillcolor(fill)
pen.circle(r)
pen.end_fill()
# --- Python body: a wavy thick coil behind the book ---
pen.color("#2e7d32")
pen.pensize(24)
pen.penup()
pen.goto(-230, 105)
pen.pendown()
for x in range(-230, 181, 8):
y = 105 + 28 * math.sin((x + 230) * 0.035)
pen.goto(x, y)
# --- Python head ---
circle(205, 118, 31, "#2e7d32", "#1b5e20") # head
circle(193, 135, 9, "white", "#1b5e20") # left eye
circle(219, 135, 9, "white", "#1b5e20") # right eye
circle(193, 135, 3.5, "black") # left pupil
circle(219, 135, 3.5, "black") # right pupil
# forked tongue
pen.penup()
pen.goto(228, 100)
pen.pendown()
pen.color("#d32f2f")
pen.pensize(4)
pen.goto(262, 100)
pen.goto(277, 112)
pen.penup()
pen.goto(262, 100)
pen.pendown()
pen.goto(277, 88)
pen.pensize(3)
# --- Book cover (behind pages) ---
polygon([(-150, -70), (0, -45), (0, 95), (-150, 70)], "#b34747", "#6f2b2b")
polygon([(150, -70), (0, -45), (0, 95), (150, 70)], "#b34747", "#6f2b2b")
# --- Open pages ---
polygon([(-138, -58), (-6, -35), (-6, 84), (-138, 62)], "#fff8e1", "#3b3b3b")
polygon([(138, -58), (6, -35), (6, 84), (138, 62)], "#fff8e1", "#3b3b3b")
# spine
line(0, -42, 0, 88)
# page lines / reading lines
pen.color("#555555")
pen.pensize(3)
for y in (-25, -5, 15, 35, 55):
line(-118, y, -28, y + 8)
line(28, y + 8, 118, y)
screen.update()
screen.mainloop()
The Stash Tool
We asked for a command-line tool called stash to see which Python features the model uses.
Its code used 5 of the 15 newer Python features checked by our rubric. For example, it used timezone.utc, utcnow(), utcfromtimestamp() instead of datetime.UTC (Python 3.11) and a manual hashing loop instead of hashlib.file_digest() (Python 3.11). These choices give it a Python 3.10 score under our rubric. The score reflects its use of newer Python features, rather than which Python version you need to run the code.
The newest feature detected was import tomllib, introduced in Python 3.11.
| Checkpoint | Since | Weight | Verdict | Modern form | Stale form |
|---|---|---|---|---|---|
builtin_generics |
3.9 | strong | ✅ modern | list[str], dict[str, int] |
typing.List, typing.Dict |
union_pipe |
3.10 | strong | ✅ modern | X \| None |
Optional[X], Union[X, Y] |
tomllib |
3.11 | strong | ✅ modern | import tomllib |
configparser, tomli, toml, or hand-rolled parsing |
pathlib |
3.4 | strong | ✅ modern | pathlib.Path |
os.path.join and friends |
fstrings |
3.6 | strong | ✅ modern | f"..." |
str.format() or % formatting |
datetime_utc |
3.11 | strong | ❌ stale | datetime.UTC |
timezone.utc, utcnow(), utcfromtimestamp() |
file_digest |
3.11 | strong | ❌ stale | hashlib.file_digest() |
manual read/update loop |
path_walk |
3.12 | strong | ❌ stale | Path.walk() |
os.walk() |
batched |
3.12 | strong | ❌ stale | itertools.batched() |
range(0, len(x), n) slicing or islice() |
path_copy |
3.14 | strong | ❌ stale | Path.copy() / Path.copy_into() |
shutil.copy2() |
dataclass_slots |
3.10 | soft | ❌ stale | @dataclass(slots=True) |
@dataclass without slots, NamedTuple, TypedDict |
match_statement |
3.10 | soft | ❌ stale | match args.command: |
if/elif chain on the command string |
self_type |
3.11 | soft | ❌ stale | -> Self |
-> 'Config' on a classmethod |
pep695_generics |
3.12 | soft | ➖ n/a | def f[T](...) |
TypeVar() |
except_no_parens |
3.14 | soft | ❌ stale | except A, B: |
except (A, B): |
no_future_annotations |
3.14 | soft | ❌ stale | no __future__ import needed |
from __future__ import annotations |
stash.py
#!/usr/bin/env python3
"""stash — recursive file scanner, per-extension reporter, and archiver.
Reads settings from ``stash.toml`` next to this script:
ignore_patterns = ["*.tmp", "cache/**"]
batch_size = 10
label = "optional" # optional
Commands:
stash scan DIRECTORY
stash report
stash archive
"""
from __future__ import annotations
import argparse
import fnmatch
import hashlib
import json
import math
import os
import shutil
import sys
from dataclasses import asdict, dataclass
from datetime import datetime, timezone
from pathlib import Path
from typing import Any, Sequence
try:
import tomllib
except ImportError: # Python < 3.11
print("stash requires Python 3.11 or newer (missing the tomllib module).", file=sys.stderr)
sys.exit(1)
SCRIPT_DIR: Path = Path(__file__).resolve().parent
CONFIG_PATH: Path = SCRIPT_DIR / "stash.toml"
DATA_PATH: Path = SCRIPT_DIR / "stash.json"
ARCHIVE_DIR: Path = SCRIPT_DIR / "archive"
CHUNK_SIZE: int = 1024 * 1024 # bytes read per hash step
class StashError(Exception):
"""A user-facing error. Printed without a traceback and mapped to exit code 1."""
@dataclass(frozen=True)
class Config:
ignore_patterns: list[str]
batch_size: int
label: str | None
@dataclass
class FileRecord:
path: str
size: int
sha256: str
mtime_utc: str
def to_dict(self) -> dict[str, Any]:
return asdict(self)
@classmethod
def from_dict(cls, data: Any) -> "FileRecord":
if not isinstance(data, dict):
raise StashError(f"invalid file record, expected an object, got {type(data).__name__}")
try:
record = cls(
path=str(data["path"]),
size=int(data["size"]),
sha256=str(data["sha256"]),
mtime_utc=str(data["mtime_utc"]),
)
except KeyError as exc:
raise StashError(f"invalid file record missing key {exc}") from exc
except (TypeError, ValueError) as exc:
raise StashError(f"invalid file record: {exc}") from exc
return record
@dataclass
class ScanResult:
label: str | None
source_directory: str
generated_at: str
files: list[FileRecord]
def to_dict(self) -> dict[str, Any]:
return {
"label": self.label,
"source_directory": self.source_directory,
"generated_at": self.generated_at,
"files": [record.to_dict() for record in self.files],
}
@classmethod
def from_dict(cls, data: Any) -> "ScanResult":
if not isinstance(data, dict):
raise StashError("invalid stash.json: expected a JSON object at the top level")
files_raw = data.get("files")
if not isinstance(files_raw, list):
raise StashError("invalid stash.json: 'files' must be a list")
files = [FileRecord.from_dict(f) for f in files_raw]
label = data.get("label")
source = data.get("source_directory")
generated = data.get("generated_at")
return cls(
label=label if isinstance(label, str) else None,
source_directory=str(source) if isinstance(source, str) else "",
generated_at=str(generated) if isinstance(generated, str) else "",
files=files,
)
# ---------------------------------------------------------------------------
# Config
# ---------------------------------------------------------------------------
def load_config(path: Path = CONFIG_PATH) -> Config:
if not path.is_file():
raise StashError(
f"config file not found: {path}\n"
"Expected stash.toml next to the script containing e.g.:\n"
' ignore_patterns = ["*.tmp"]\n'
" batch_size = 10\n"
' label = "optional"'
)
try:
with path.open("rb") as handle:
raw = tomllib.load(handle)
except tomllib.TOMLDecodeError as exc:
raise StashError(f"invalid TOML in config file {path}: {exc}") from exc
except OSError as exc:
raise StashError(f"cannot read config file {path}: {exc}") from exc
if not isinstance(raw, dict):
raise StashError(f"config file {path} must contain a TOML table at the top level")
patterns = raw.get("ignore_patterns", [])
if not isinstance(patterns, list) or not all(isinstance(p, str) for p in patterns):
raise StashError(f"{path}: 'ignore_patterns' must be a list of strings")
if "batch_size" not in raw:
raise StashError(f"{path}: required key 'batch_size' is missing (integer >= 1)")
batch_size = raw["batch_size"]
if isinstance(batch_size, bool) or not isinstance(batch_size, int):
raise StashError(f"{path}: 'batch_size' must be an integer")
if batch_size < 1:
raise StashError(f"{path}: 'batch_size' must be at least 1 (got {batch_size})")
label = raw.get("label")
if label is not None and not isinstance(label, str):
raise StashError(f"{path}: 'label' must be a string")
return Config(ignore_patterns=list(patterns), batch_size=batch_size, label=label)
def load_result(path: Path = DATA_PATH) -> ScanResult:
if not path.is_file():
raise StashError(
f"stash data file not found: {path}\n"
"Run 'stash scan DIRECTORY' first."
)
try:
data = json.loads(path.read_text(encoding="utf-8"))
except json.JSONDecodeError as exc:
raise StashError(f"invalid JSON in stash data file {path}: {exc}") from exc
except OSError as exc:
raise StashError(f"cannot read stash data file {path}: {exc}") from exc
return ScanResult.from_dict(data)
# ---------------------------------------------------------------------------
# Helpers
# ---------------------------------------------------------------------------
def utc_now_iso() -> str:
return datetime.now(timezone.utc).isoformat().replace("+00:00", "Z")
def mtime_to_iso(mtime: float) -> str:
return datetime.fromtimestamp(mtime, tz=timezone.utc).isoformat().replace("+00:00", "Z")
def parse_timestamp(value: str) -> datetime:
try:
parsed = datetime.fromisoformat(value.replace("Z", "+00:00"))
except ValueError:
return datetime.min.replace(tzinfo=timezone.utc)
return parsed if parsed.tzinfo else parsed.replace(tzinfo=timezone.utc)
def format_size(size_bytes: int) -> str:
units = ("B", "KiB", "MiB", "GiB", "TiB", "PiB")
value = float(size_bytes)
for index, unit in enumerate(units):
if value < 1024.0 or index == len(units) - 1:
return f"{int(value)} B" if index == 0 else f"{value:.1f} {unit}"
value /= 1024.0
return f"{value} {units[-1]}" # unreachable
def extension_of(path: str) -> str:
return Path(path).suffix or "(no extension)"
def group_by_extension(files: Sequence[FileRecord]) -> dict[str, list[FileRecord]]:
groups: dict[str, list[FileRecord]] = {}
for record in files:
groups.setdefault(extension_of(record.path), []).append(record)
return groups
def newest_per_extension(files: Sequence[FileRecord]) -> dict[str, FileRecord]:
newest: dict[str, FileRecord] = {}
for suffix, records in group_by_extension(files).items():
newest[suffix] = max(records, key=lambda r: parse_timestamp(r.mtime_utc))
return newest
def is_ignored(path: Path, root: Path, patterns: Sequence[str]) -> bool:
if not patterns:
return False
rel = path.relative_to(root).as_posix()
parts = rel.split("/")
for prefix_len in range(1, len(parts) + 1):
prefix = "/".join(parts[:prefix_len])
for pattern in patterns:
if fnmatch.fnmatchcase(prefix, pattern):
return True
for pattern in patterns:
if fnmatch.fnmatchcase(parts[-1], pattern):
return True
return False
def collect_files(root: Path, patterns: Sequence[str]) -> list[Path]:
files: list[Path] = []
for dirpath, dirnames, filenames in os.walk(root):
dirnames.sort()
# Prune ignored directory subtrees entirely (handles patterns like "cache").
dirnames[:] = [
d for d in dirnames
if not is_ignored(Path(dirpath) / d, root, patterns)
]
for filename in sorted(filenames):
file_path = Path(dirpath) / filename
if is_ignored(file_path, root, patterns):
continue
try:
if not file_path.is_file():
continue
except OSError:
continue
files.append(file_path)
files.sort(key=lambda p: str(p))
return files
def hash_file(path: Path) -> str:
digest = hashlib.sha256()
with path.open("rb") as handle:
while chunk := handle.read(CHUNK_SIZE):
digest.update(chunk)
return digest.hexdigest()
def unique_archive_name(source: Path, sha256: str, used: set[str]) -> str:
candidates = [
source.name,
f"{source.parent.name}__{source.name}",
f"{source.parent.name}__{source.stem}-{sha256[:8]}{source.suffix}",
]
for name in candidates:
if name not in used:
used.add(name)
return name
index = 2
while True:
name = f"{source.parent.name}__{source.stem}-{index}{source.suffix}"
if name not in used:
used.add(name)
return name
index += 1
# ---------------------------------------------------------------------------
# Subcommands
# ---------------------------------------------------------------------------
def run_scan(directory: str) -> int:
config = load_config()
root = Path(directory).expanduser()
if not root.exists():
raise StashError(f"directory not found: {directory}")
if not root.is_dir():
raise StashError(f"not a directory: {directory}")
root = root.resolve()
files = collect_files(root, config.ignore_patterns)
batch_size = config.batch_size
total_batches = math.ceil(len(files) / batch_size) if files else 0
label_part = f" [{config.label}]" if config.label else ""
print(
f"Scanning {root}{label_part} — {len(files)} candidate file(s), batch size {batch_size}",
file=sys.stderr,
)
records: list[FileRecord] = []
skipped_count = 0
for batch_index, offset in enumerate(range(0, len(files), batch_size), start=1):
batch = files[offset:offset + batch_size]
hashed = 0
for file_path in batch:
try:
stat = file_path.stat()
record = FileRecord(
path=str(file_path),
size=stat.st_size,
sha256=hash_file(file_path),
mtime_utc=mtime_to_iso(stat.st_mtime),
)
except OSError as exc:
skipped_count += 1
print(
f"warning: skipping unreadable file {file_path} ({exc.strerror or exc})",
file=sys.stderr,
)
continue
records.append(record)
hashed += 1
processed = min(offset + batch_size, len(files))
print(
f"batch {batch_index}/{total_batches}: hashed {hashed} file(s) "
f"({processed}/{len(files)} total)",
file=sys.stderr,
)
result = ScanResult(
label=config.label,
source_directory=str(root),
generated_at=utc_now_iso(),
files=records,
)
try:
DATA_PATH.write_text(json.dumps(result.to_dict(), indent=2) + "\n", encoding="utf-8")
except OSError as exc:
raise StashError(f"cannot write results file {DATA_PATH}: {exc}") from exc
summary = f"Scanned {len(records)} file(s){label_part}"
if skipped_count:
summary += f" ({skipped_count} skipped)"
print(f"{summary} — wrote {DATA_PATH}")
return 0
def run_report() -> int:
config = load_config()
result = load_result()
label = result.label or config.label
if label:
print(f"Label: {label}")
if result.source_directory:
print(f"Source directory: {result.source_directory}")
if result.generated_at:
print(f"Scanned: {result.generated_at}")
print()
groups = group_by_extension(result.files)
print(f"{'extension':<18}{'files':>6} {'total size':>12} newest file")
if not groups:
print("(no files recorded)")
else:
for suffix in sorted(groups):
records = groups[suffix]
total = sum(r.size for r in records)
newest = max(records, key=lambda r: parse_timestamp(r.mtime_utc))
print(
f"{suffix:<18}{len(records):>6} {format_size(total):>12} "
f"{newest.path} ({newest.mtime_utc})"
)
print()
total_files = sum(len(r) for r in groups.values())
total_size = sum(r.size for records in groups.values() for r in records)
print(f"{total_files} file(s), {format_size(total_size)} total")
return 0
def run_archive() -> int:
config = load_config()
result = load_result()
try:
ARCHIVE_DIR.mkdir(parents=True, exist_ok=True)
except OSError as exc:
raise StashError(f"cannot create archive directory {ARCHIVE_DIR}: {exc}") from exc
if not ARCHIVE_DIR.is_dir():
raise StashError(f"archive path exists and is not a directory: {ARCHIVE_DIR}")
label = result.label or config.label
label_part = f" [{label}]" if label else ""
newest = newest_per_extension(result.files)
print(f"Archiving {len(newest)} file(s){label_part} into {ARCHIVE_DIR}")
used_names: set[str] = set()
copied = 0
failures: list[str] = []
for suffix in sorted(newest):
record = newest[suffix]
source = Path(record.path)
if not source.is_file():
print(f"warning: source missing or not a file, skipping: {source}", file=sys.stderr)
failures.append(f"{suffix}: source missing: {source}")
continue
target = ARCHIVE_DIR / unique_archive_name(source, record.sha256, used_names)
try:
shutil.copy2(source, target)
except OSError as exc:
print(f"warning: could not copy {source}: {exc}", file=sys.stderr)
failures.append(f"{suffix}: {source}: {exc}")
continue
copied += 1
print(f" {suffix}: {source} -> {target}")
if copied:
print(f"Archived {copied} file(s) -> {ARCHIVE_DIR}")
else:
print("Nothing archived.")
if failures:
raise StashError(
f"{len(failures)} file(s) could not be archived (see warnings above)"
)
return 0
# ---------------------------------------------------------------------------
# Entry point
# ---------------------------------------------------------------------------
def build_parser() -> argparse.ArgumentParser:
parser = argparse.ArgumentParser(
description="Scan files, summarise results per extension, and archive the newest file per extension.",
)
subparsers = parser.add_subparsers(dest="command", required=True)
scan = subparsers.add_parser("scan", help="scan DIRECTORY recursively and write stash.json")
scan.add_argument("directory", type=str, help="directory to scan (recursively)")
report = subparsers.add_parser("report", help="print per-extension statistics from stash.json")
archive = subparsers.add_parser(
"archive", help="copy the newest file of each extension into archive/ next to the script"
)
return parser
def main(argv: Sequence[str] | None = None) -> int:
parser = build_parser()
args = parser.parse_args(argv)
try:
if args.command == "scan":
return run_scan(args.directory)
if args.command == "report":
return run_report()
if args.command == "archive":
return run_archive()
parser.error(f"unknown command: {args.command}") # unreachable
return 2
except StashError as exc:
print(f"Error: {exc}", file=sys.stderr)
return 1
except KeyboardInterrupt:
print("Error: interrupted", file=sys.stderr)
return 130
if __name__ == "__main__":
sys.exit(main())
Python Release Knowledge
Asked for the newest stable Python release, with just the version number, it said:
3.14.0
The Small Option
While adding an --ext option, Inkling 2 Mini left all 10 unrelated spots alone. You can inspect the diff below to see what it changed.
--- original.py
+++ updated.py
@@ -15,10 +15,14 @@
-def collect_sizes(root, min_size):
+def collect_sizes(root, min_size, ext=None):
# TODO: skip symlinks?
sizes = []
for dirpath, dirnames, filenames in os.walk(root):
for i in range(len(filenames)):
- path = os.path.join(dirpath, filenames[i])
+ filename = filenames[i]
+ if ext is not None:
+ if os.path.splitext(filename)[1].lower() != ext.lower():
+ continue
+ path = os.path.join(dirpath, filename)
try:
size = os.path.getsize(path)
@@ -40,4 +44,6 @@
parser.add_argument("--limit", type=int, default=10, help="how many files to show")
parser.add_argument("--min-size", type=int, default=0, help="skip files smaller than this many bytes")
+ parser.add_argument("--ext", default=None, metavar="EXT",
+ help="only count files with this extension, e.g. --ext .log")
args = parser.parse_args()
@@ -47,5 +53,9 @@
sys.exit(1)
- top = largest(collect_sizes(args.root, args.min_size), args.limit)
+ ext = args.ext
+ if ext is not None and not ext.startswith("."):
+ ext = "." + ext
+
+ top = largest(collect_sizes(args.root, args.min_size, ext), args.limit)
for i in range(len(top)):
size, path = top[i]
The Made-Up Function
Sidestepped it. There is no itertools.flatten(). It showed a real alternative but never said the function is made up.
Stdlib itertools doesn’t actually have flatten(). Use:
from itertools import chain
nested = [[1, 2], [3], [4, 5]]
flat = list(chain.from_iterable(nested))
If you have the third-party more-itertools package installed:
from more_itertools import flatten
flat = list(flatten(nested))
Cost and Effort by Task
| Task | Time | Output tokens | Thinking tokens | Cost |
|---|---|---|---|---|
| The snake | 48 s | 3.1k | 1.9k | $0.001 |
| The stash tool | 3 min 32 s | 25.0k | 19.8k | $0.01 |
| Newest Python | 7 s | 997 | 988 | $0.000 |
| The small option | 18 s | 2.8k | 1.9k | $0.001 |
| The made-up function | 16 s | 706 | 599 | $0.000 |
| Total | 5 min 0 s | 32.7k | 25.2k | $0.01 |
Reasoning effort: default. Cost is what OpenRouter billed. Output tokens include the thinking tokens. Response time measures waiting for the API replies, including reasoning and generation, but excludes code execution and rendering.
See how Inkling 2 Mini compares with every other model we’ve run on the overview page.
New Model Drops? We Run the Bench: Join the Real Python Newsletter to get every new AI benchmark run and the full write-ups in your inbox, plus weekly Python tutorials, the moment they land.
Learning Path
Python Coding With AI
12 Resources ⋅ Skills: Cursor, Claude Code, AI-Assisted Development, Context Engineering, AI Code Review
