Inkling-Small

Thinking Machines Lab·Pending·Released 2026-07-30textvisionaudio
Context1,000k
Max output-
Input / 1M$0.3 ↗
Output / 1M$1.2 ↗
Cached in$0.06
Benchmarks29

Benchmarks

Cited public sources · rank is exact-benchmark, apples-to-apples

Chess Puzzles (Epoch AI run)#61/9218% ↗
Mystery Game Puzzles (Epoch AI run)#66/686% ↗
SAGE (Vals AI)#66/7732.7% ↗
Terminal-Bench 4.0 (Vals AI)#39/400% ↗
LMArena Elo#100/1411405.2 ↗
AIME 2026#3/595.5% ↗
ARC-AGI-1#1/284% ↗
ARC-AGI-2#3/640.1% ↗
BrowseComp (with context management)77.4% ↗
CharXiv RQ#2/277.4% ↗
CharXiv RQ (with python)#2/281.3% ↗
HMMT Feb 202690.2% ↗
IFBench#1/1282.2% ↗
MCP Atlas (all)79.2% ↗
MMAU#2/277% ↗
Terminal-Bench 2.1 (best harness)#1/264.7% ↗
Toolathlon Verified#5/654.4% ↗
Toolathlon-Verified (HKUST)#17/2354.4% ↗

Inkling-Small FAQ

What is Inkling-Small?
Inkling-Small is a large language model from Thinking Machines Lab, released on July 30, 2026. It accepts text, image and audio input.
When was Inkling-Small released?
Thinking Machines Lab released Inkling-Small on July 30, 2026.
How much does Inkling-Small cost?
Inkling-Small costs $0.30 per 1M input tokens and $1.20 per 1M output tokens through the Thinking Machines Lab API. Cached input costs $0.06 per 1M tokens. Prices are from Thinking Machines Lab's official pricing as of October 5, 2026.
What is Inkling-Small's context window?
Inkling-Small has a 1M-token context window.
How does Inkling-Small score on benchmarks?
Inkling-Small has 29 cited benchmark scores, including Chess Puzzles (Epoch AI run) 18%, FrontierMath Tier 4 v2 (Epoch AI run) 17.1% and FrontierMath Tiers 1-3 v2 (Epoch AI run) 46.3%. Each score links to its source.

Access Inkling-Small and every other model through one endpoint with automatic failover: Respan gateway.