메인 콘텐츠로 건너뛰기
AITier

Inkling Small

Thinking Machines오픈 웨이트Apache 2.0 · 상업적 사용 가능

설명

Inkling-Small is Thinking Machines Lab's efficient open-weights MoE multimodal model (276B total / 12B active parameters) released under Apache 2.0. It accepts text, image, and audio inputs and generates text, with native reasoning, variable thinking effort, and a context window up to 1M tokens (Tinker exposes 64K and 256K configurations). Hugging Face weights: thinkingmachines/Inkling-Small and thinkingmachines/Inkling-Small-NVFP4. Vendor self-host VRAM: BF16 ≥ ~600 GB aggregated; NVFP4 ≥ ~180 GB aggregated. SWE-Bench Verified 80.2% vs Inkling 77.6% (same bash-only harness). Fine-tuning and playground chat are available via Tinker. Official Tinker serverless inference (256K, list): $0.30 / $1.20 per 1M input/output tokens ($0.06 cached).

출시일
2026-07-30
파라미터
276.0B
컨텍스트 길이
262K
모달리티
audio, image, text

능력 레이더

27
general
52
coding
90
reasoning
64
science
50
agents
80
multimodal

랭킹

도메인#순위점수소스
에이전트형 역량60
48.0
LS
코딩 랭킹108
72.0
AA
종합 랭킹237
42.0
AA
멀티모달 랭킹91
49.0
LS
과학73
72.0
AA

벤치마크 점수 (LLM Stats)

(LLM Stats (zeroeval))

Agents

AA-Briefcase917.00 / 3000자체 보고
Toolathlon54.4%자체 보고
Tau3 Banking15.5%자체 보고

Audio

MMAU77.0%자체 보고

Chat

VoiceBench Avg90.1%자체 보고

Factuality

SimpleQA Verified20.6%자체 보고

General

GDPval-AA1269.00 / 3000자체 보고
Global-MMLU-Lite86.7%자체 보고
Artificial Analysis40.0%자체 보고

Instruction Following

IFBench82.2%자체 보고

Math

AIME 202695.5%자체 보고

Reasoning

GPQANYU + Cohere + Anthropic (2023)89.5%자체 보고
ARC-AGI84.0%자체 보고
SWE-Bench Verified80.2%자체 보고
MCP Atlas79.6%자체 보고
CharXiv-R77.4%자체 보고
BrowseCompOpenAI (2025)77.4%자체 보고
Terminal-Bench 2.164.7%자체 보고
SWE-Bench ProPrinceton NLP (2024)55.9%자체 보고
SciCode48.7%자체 보고
ARC-AGI v240.1%자체 보고
Humanity's Last Exam31.6%자체 보고
CritPT8.3%자체 보고

Vision

MMMU-Pro74.0%자체 보고

AA 평가 지수

(Artificial Analysis)
Gpqa(NYU + Cohere + Anthropic (2023))
89.5
Lcr(Artificial Analysis)
75.7
Terminalbench V2 1
55.1
Coding Index(Artificial Analysis)
52.9
Scicode(UIUC + Argonne National Lab (2024))
49.7
Hle(Center for AI Safety + Scale AI (2025))
33.3
Intelligence Index(Artificial Analysis)
25.7
Tau Banking
18.8
Terminalbench V4 0
1.0

LLM Stats 카테고리 점수

(LLM Stats (zeroeval))
Legal
100
Finance
100
Productivity
100
Agents
100
Reasoning
100
General
92
Chat
90
Language
90
Safety
90
Speech To Text
90
Communication
90
Instruction Following
80
Multimodal
80
Search
80
Frontend Development
80
Audio
80
Physics
70
Biology
70
Chemistry
70
Code
70
Spatial Reasoning
60
Vision
60
Math
50
Tool Calling
50
Factuality
20

가격

입력 가격$0.3 / 1M 토큰
출력 가격$1.2 / 1M 토큰
혼합 가격 (3:1)$0.525 / 1M 토큰
캐시 읽기 가격$0.116 / 1M 토큰

속도

토큰/초222.6
첫 토큰 지연1.78s
첫 응답 지연10.77s

공급자 가격 순위

공급자 가격 순위

13개 공급자

최저가: Thinking Machines Lab최고가: AI/ML API
공급자입력출력
1Thinking Machines Lab최저가
$0
$0
2DeepInfra
$0
$0
3Thinking Machines주요
$0.3
$1.2
4NanoGPT
$0.45
$1.2
5OpenRouter
$0.45
$1.2
6Deep Infra
$0.45
$1.2
7Kilo Gateway
$0.45
$1.2
8Vercel AI Gateway
$0.45
$1.2
9Hugging Face
$0.5
$1.2
10Arcee
$0.5
$1.2
11CommandCode
$0.5
$1.2
12LLMTR
$0.58
$1.44
13AI/ML API
$0.61364
$1.52352

이 모델의 다양한 API 공급자 간 가격 비교.

외부 링크

LLM StatsArtificial Analysis