Inkling Small

Tool callingReasoningImage inputAudio input

Hugging Face Launch post Website

Multimodal MoE reasoning model (276B total, 12B active) for text, image, and audio

Context: 524K
Released: 2mo ago
Parameters: —
Input / 1M tokens: $0.45
Output / 1M tokens: $1.20
Cached input / 1M tokens: $0.10

Sources:models.devOpenRouter

Benchmarks

DeepSWE is the primary coding benchmark where an exact model evaluation exists. Other sourced evaluations appear below; missing scores are not zero.

Not evaluated in DeepSWE 1.1 as of 2026-09-22. Results are only shown for exact model versions in the official dataset. Official leaderboard.

Other published benchmarks

Evaluation	Reported score and source
ARC-AGI-2	33.1% Third-party evaluator · inkling-small_high Reasoning effort: high External evaluation via Epoch AI Original-source terms apply
ARC-AGI-2	4.9% Third-party evaluator · inkling-small_low Reasoning effort: low External evaluation via Epoch AI Original-source terms apply
ARC-AGI-2	13.6% Third-party evaluator · inkling-small_medium Reasoning effort: medium External evaluation via Epoch AI Original-source terms apply
ARC-AGI-2	4.0% Third-party evaluator · inkling-small_minimal Reasoning effort: minimal External evaluation via Epoch AI Original-source terms apply
ARC-AGI-2	0.0% Third-party evaluator · inkling-small_none Reasoning effort: none External evaluation via Epoch AI Original-source terms apply
ARC-AGI-2	40.1% Third-party evaluator · inkling-small_xhigh Reasoning effort: xhigh External evaluation via Epoch AI Original-source terms apply
FrontierMath-Tier-4-v2-Private	17.1% Epoch AI · inkling-small_xhigh Reasoning effort: xhigh Epoch AI evaluation
FrontierMath-Tiers-1-3-v2-Private	46.3% Epoch AI · inkling-small_xhigh Reasoning effort: xhigh Epoch AI evaluation
GPQA diamond	88.5% Epoch AI · inkling-small_xhigh Reasoning effort: xhigh Epoch AI evaluation
OTIS Mock AIME 2024-2025	90.0% Epoch AI · inkling-small_xhigh Reasoning effort: xhigh Epoch AI evaluation
SimpleQA Verified	19.1% Epoch AI · inkling-small_xhigh Reasoning effort: xhigh Epoch AI evaluation

Benchmark compilation: Epoch AI, Capabilities & benchmarking (CC BY attribution). Scores are converted to percentages and model names matched to the Vertical Tabs catalog. Individual evaluations are credited to their original sources. External results may carry separate licences and evaluation configurations; do not interpret scores from different harnesses as directly comparable.

Specifications

Developer: Thinkingmachines
Model family: —
Type: Language model
Predecessor: —
Successor: —
Released: July 30, 2026
Model ID: thinkingmachines/inkling-small
Openness: Open weights
License: Apache-2.0
Parameters: —
Input modalities: text, image, audio
Output modalities: text
Modality: text+image+audio->text
Tokenizer: Other
Max output tokens: 262K
Hugging Face: thinkingmachines/Inkling-Small
Source freshness: Metadata synced Oct 8, 2026, 10:17 PM UTC

Additional source evidence

Source model: thinkingmachines/inkling-small
Source family: ling
Last updated: 2026-07-30
Model context: 1.0M
Model output limit: 1.0M
Open weights: Yes
File attachments: Yes
Reasoning: Yes
Tool calling: Yes
Temperature control: Yes

Downloadable weights

Hugging Face

Complete source record (JSON)

{
  "id": "thinkingmachines/inkling-small",
  "name": "Inkling Small",
  "description": "Multimodal MoE reasoning model (276B total, 12B active) for text, image, and audio",
  "family": "ling",
  "last_updated": "2026-07-30",
  "attachment": true,
  "reasoning": true,
  "tool_call": true,
  "temperature": true,
  "modalities": {
    "input": [
      "text",
      "image",
      "audio"
    ],
    "output": [
      "text"
    ]
  },
  "limit": {
    "context": 1048576,
    "output": 1048576
  },
  "release_date": "2026-07-30",
  "open_weights": true,
  "license": "Apache-2.0",
  "weights": [
    {
      "label": "Hugging Face",
      "url": "https://huggingface.co/thinkingmachines/Inkling-Small"
    }
  ]
}

Infrastructure

Provider availability

All providers

Provider	Endpoint	Context	Input /M tokens	Output /M tokens	Cached input /M tokens	Cache write /M tokens	Uptime 1d	Data handling
DeepInfra	deepinfra/fp8	524K	$0.45	$1.20	$0.10	—	100.00%	ZDR: Yes Training: Unknown Prompt logging: Unknown Retention: Unknown ZDR source Privacy policy Checked 2026-10-15

Provider data:OpenRouter

Deployment graph

Can I run this?

Vertical Tabs only shows compatibility when it has structured evidence tying this exact model to hardware, runtime, quantization, memory use, and observed performance.

Browse hardware