systemcards.org

Inkling 2 Small Model Card

Publisher
thinking_machines first-party
Type
model_card
Models
Inkling 2 Small
Published
2026-11-05
Status
active
Safety evals
yes
Risk domains
CBRN Cyber offence Loss of control Societal harms
Openness
closed weights
Source
https://thinkingmachines.ai/model-card/inkling-2-small/
Related
announcement https://thinkingmachines.ai/news/inkling-2-small/
weights https://huggingface.co/thinkingmachines/Inkling-2-Small
First seen
2026-11-06T06:26:57Z
Versions
2
Notes
At launch the model is closed-weight (Tinker API only). The announcement says full weights are intended under Apache 2.0 'next week' after external safety testing, so openness should move to open_weight_permissive once the weights ship. The model card's License field is blank.

Report an issue with this entry

Version history

FetchedContent typeSizeFingerprintOriginalWhat changed
2026-11-13T06:20:19Z text/html 332,488 0309a125b7a1376f archived copy Extractor drift: 1 line(s) missing from the extracted text are still present in the raw capture; not a publisher deletion.
2026-11-06T06:26:57Z text/html 269,522 5a926f84f5a08485 archived copy initial version

Provenance

Show 4 changelog entries (when, how, and why this entry changed)
  1. 2026-11-06T06:26:57Z add run 2026-11-06T06:18Z-local · agent · lead: phase_a
    Thinking Machines Lab's own model card for Inkling 2 Small, its second-generation model (594B total / 17B active MoE, text+image input, 1M context), released 2026-11-05 alongside an announcement post. This is a new generation, not a variant of the catalogued Inkling or Inkling-Small. The card includes a full eval table with a Safety block (DistressBench, FORTRESS adversarial/benign, StrongREJECT) and a Cybersecurity block (CyberBench, PatchEval). It also has a Safety section describing CBRN and cyber uplift testing, including refusal-suppressed variants, plus loss-of-control evals (agentic capability, strategic deception, sabotage). The companion announcement reports alignment results: internal misbehavior, covert misbehavior and destructive-action rates, Goodhart Labs HoneyBench reward hacking, and Apollo Research covert-action evals.
    Criteria: document_retrievable=pass; min_publication_date=pass; publisher_on_allowlist=pass · attested about_a_specific_model_or_eval=true · attested covered_model_class=true · attested distinct_model_release=true · attested notable_release=true · attested primary_source=true
  2. 2026-11-13T06:20:19Z new_version run 2026-11-13T06:17Z-local · monitor · lead: monitor
    Content fingerprint changed on scheduled re-fetch
  3. 2026-11-13T06:20:19Z field_update run 2026-11-13T06:17Z-local · monitor
    change_summary: null → "Extractor drift: 1 line(s) missing from the extracted text are still present in the raw capture; not a publisher deletion."
    Deleted lines found verbatim in the new raw capture
  4. 2026-11-16T06:26:36Z field_update run 2026-11-16T06:18Z-local · agent · lead: agent_search
    related_urls: [{"kind": "announcement", "url": "https://thinkingmachines.ai/news/inkling-2-small/"}] → [{"kind": "announcement", "url": "https://thinkingmachines.ai/news/inkling-2-small/"}, {"kind": "weights", "url": "https://huggingface.co/thinkingmachines/Inkling-2-Small"}]
    The model card links the official Hugging Face weights repo (NVFP4 main checkpoint; MXFP8 variant linked from it), which carries the same card text. Adding it as the weights companion; existing announcement link kept.

Extracted text

Show extracted text (from the latest fetched version; formatting is lossy — use the original for reading)
Inkling 2 Small Model Card
1. General information
Model name
Inkling 2 Small
Legal name of the model provider
Thinking Machines Lab, Inc.
Date of release
Nov 5, 2026
License
Apache 2.0
Intended uses
Inkling 2 Small is a general-purpose multimodal model that accepts text and image inputs and generates text outputs. It is intended for use in English and other languages, and across multiple coding languages. The model is designed to be used by developers building AI-powered applications, including agentic and tool-use systems, coding assistants, chatbots, and retrieval-augmented generation systems, and is suitable for general-purpose conversational use, instruction-following, and other natural language and visual tasks. It is released with open weights to support research, fine-tuning and integration into third-party products by downstream developers.
Languages
English, general multilingual support
2. Model properties
Model type
Multimodal autoregressive transformer
Architecture type
A 48 layer decoder-only transformer with a sparse Mixture-of-Experts (MoE) feed-forward backbone: each token is routed to 8 of 384 experts. Attention is a hybrid of local and global layers. The model is natively multimodal – images are encoded via a vision transformer – with image and text inputs projected into a shared hidden space and processed jointly by the decoder.
Parameters
594B total parameters, 17B active
Numerics support
MXFP8 and NVFP4
Input modalities
- Text: UTF-8 encoded text
- Image: Any pixel-based image input. For optimal performance, each image dimension should be between 40px to 4096px.
Inkling 2 Small supports a context window of up to 1M tokens.
Output modalities
Text: UTF-8 encoded text
3. Methods of distribution
The weights are available for download through Hugging Face here and here for MXFP8. Developers can also access the model using the Tinker fine-tuning platform, which provides API access.
The model is also available via API access through third party inference providers.
API access to the model
- For accessing our model via Tinker, you can get started by referring to the Tinker Cookbook here, and installing the tml-renderers package here.
- Tinker playground also provides inference-only access to the model here.
- For accessing our model via third party inference providers, you can refer to the documentation from our inference provider partners.
Running the open weights model on your own system
Required hardware
The model is distributed in two checkpoint formats.
The main NVFP4 checkpoint requires a GPU cluster with at least 500GB of aggregated VRAM. This can be met with either of the following example configurations:
- 2x NVIDIA B300 GPUs
- 4x NVIDIA H200 GPUs
The alternative MXFP8 checkpoint requires a GPU cluster with at least 800GB of aggregated VRAM. This can be met with either of the following example configurations:
- 4x NVIDIA B300 GPUs
- 8x NVIDIA H200 GPUs
The model was pretrained, midtrained and post-trained in NVFP4, so this is the native precision for the model. We recommend the NVFP4 checkpoint for all use-cases, and we’re confident that it’s the best option. The MXFP8 has a very marginal 1–2 percentage points gain over the NVFP4 checkpoint over evals, but we do not think it is meaningful for any practical use-cases.
Required software
Running the model directly on GPU hardware requires an inference deployment framework–either SGLang, vLLM, TokenSpeed, Unsloth, or Hugging Face, along with all of their respective dependency libraries.
4. Training
Data types
Training data includes a broad variety of content types, including text, images, video.
Data provenance
Training data for the model was drawn from publicly available sources, acquired from third-parties, generated internally, or synthetically generated or augmented. Publicly available data includes content from the public internet and publicly accessible repositories.
Curation methodologies
Data curation includes cleaning, processing, and modifying datasets. These processing steps, which vary by data type, may include deduplication and filtering to remove junk or other low-quality data, or to advance safety or other objectives.
5. Evaluations
| Open Weights Flash |  |  |  |  | Open Weights Pro |  |  |  |  | Closed Weights Flash |  |  |  |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Inkling 2 Small | GLM-5.3-Flash | MiMo V2.6 Flash | MiniMax-M3.1-Flash | DeepSeek-V4.1-Flash-1017 | Muse Spark Open | GLM 5.3 | MiMo V2.6 Pro | Mistral Large 4 | Kimi K3 | GPT-6 Luna | Gemini 3.8 Flash | Claude Haiku 5.5 |  |
| Model Info |  |  |  |  |  |  |  |  |  |  |  |  |  |
| Params (B)activated / total | 17/594 | 18/320 | 15/309 | 23/428 | 16/552 | 42/816 | 40/753 | 42/1020 | 49/1050 | 104/2800 | – | – | – |
| Third Party Index |  |  |  |  |  |  |  |  |  |  |  |  |  |
| AA Index (v5.0) | 30.6 | 27.8 | 26.7 | 29.1 | 32.6 (Second-best reported score in this row) | 24.9 | 31.7 | 29.9 | 26.1 | – | 27.9 | 30.8 | 36.5 (Best reported score in this row) |
| Vals Index (v2.2) | 56.1% | 50.9% | 54.8% | 53.8% | 61.3% (Best reported score in this row) | 48.6% | 59.7% (Second-best reported score in this row) | 57.8% | 56.5% | 53.0% | 52.3% | 56.0% | 58.7% |
| Coding |  |  |  |  |  |  |  |  |  |  |  |  |  |
| FrontierCode v2.0stirrup v0.2.1 | 25.4% | 17.6% | 21.6% | 20.8% | 28.5% (Second-best reported score in this row) | 7.6% | 21.6% | 24.0% | 15.0% | 17.3% | 25.2% | 18.7% | 34.9% (Best reported score in this row) |
| DeepSWE v1.1mini-swe-agent v2.4.6 | 72.1% | 63.4% | 67.9% | 73.7% | 77.0% (Best reported score in this row) | 60.0% | 69.0% | 71.9% | 61.7% | 68.5% | 66.6% | 73.8% (Second-best reported score in this row) | 59.9% |
| Terminal Bench Science v0.1mini-swe-agent v2.4.6 | 5.7% | 4.8% | 4.3% | 3.3% | 12.9% (Second-best reported score in this row) | 1.9% | 9.5% | 5.7% | 4.3% | 8.1% | 8.6% | 10.0% | 20.0% (Best reported score in this row) |
| Terminal Bench v4.0mini-swe-agent v2.4.6 | 13.1% | 32.8% | 22.7% | 25.8% | 33.8% | 8.1% | 41.9% (Best reported score in this row) | 34.9% (Second-best reported score in this row) | 24.8% | 12.6% | 12.6% | 19.7% | 32.8% |
| Terminal Bench v2.1terminus v2 | 81.3% | 84.3% | 76.4% | 81.7% | 78.4% | 76.4% | 83.9% | 87.4% (Second-best reported score in this row) | 79.5% | 85.0% | 73.0% | 87.6% (Best reported score in this row) | 79.5% |
| Vibe Code Bench v1.1openhands (vals) | 81.6% | 30.8% | 79.0% | 73.9% | 87.7% (Second-best reported score in this row) | 76.2% | 78.1% | 85.2% | 81.6% | 85.0% | 81.7% | 78.7% | 90.4% (Best reported score in this row) |
| SWE-Marathon v1.1Inkling 2 Small in Claude Code | 38.8% | 38.8% | – | – | 40.0% | – | 42.5% | – | – | 48.1% (Second-best reported score in this row) | 28.1% | – | 50.0% (Best reported score in this row) |
| SWE-Atlas-QnA v1.1mini-swe-agent v2.4.6 | 74.2% (Best reported score in this row) | – | – | – | 71.0% (Second-best reported score in this row) | 50.0% | – | – | – | 65.3% | 50.8% | 45.2% | 62.1% |
| SWE-Atlas-TW v1.1mini-swe-agent v2.4.6 | 58.1% | – | – | – | 61.1% | 41.1% | – | – | – | 63.3% (Second-best reported score in this row) | 54.4% | 71.1% (Best reported score in this row) | 63.3% (Second-best reported score in this row) |
| SWE-Atlas-Refactor v1.1mini-swe-agent v2.4.6 | 53.3% | – | – | – | 54.3% (Second-best reported score in this row) | 35.7% | – | – | – | 54.3% (Second-best reported score in this row) | 47.1% | 55.7% (Best reported score in this row) | 52.9% |
| SWEBench Multilingual | 81.2% | 88.9% (Second-best reported score in this row) | 79.9% | 79.9% | 82.9% | 77.5% | 87.6% | 82.4% | 79.7% | 91.4% (Best reported score in this row) | 78.2% | 85.8% | 83.7% |
| SWEBench Multimodal v2mini-swe-agent (vals) | 43.1% | – | – | – | – | – | – | 41.3% | – | 44.2% (Best reported score in this row) | 37.5% | 40.8% | 44.0% (Second-best reported score in this row) |
| SWEBench Pro v2 (Hard)mini-swe-agent v2.4.6 | 86.3% | 82.8% | 89.7% (Best reported score in this row) | 84.3% | 89.7% (Best reported score in this row) | 82.4% | 84.3% | 89.2% | 83.8% | 88.2% | 88.7% | 58.8% | 52.9% |
| SWEBench Pro v2 (Full)mini-swe-agent v2.4.6 | 97.7% (Best reported score in this row) | – | – | 96.6% | – | 95.0% | 95.6% | – | – | 97.7% (Best reported score in this row) | – | 94.9% | 87.2% |
| Sudo L7 | 14.1% (Best reported score in this row) | – | – | – | – | – | 9.4% | – | 13.9% (Second-best reported score in this row) | 12.2% | – | 10.6% | – |
| Autoresearch |  |  |  |  |  |  |  |  |  |  |  |  |  |
| PostTrainBench v1.2Inkling 2 Small in Claude Code | 31.9% | 36.8% (Best reported score in this row) | – | – | 30.5% | – | 34.5% (Second-best reported score in this row) | – | – | 31.5% | – | – | – |
| PMPP-Hard (AUC, 0–10h)pi v0.83.0 | 494.27 (Second-best reported score in this row) | – | – | – | 497.36 (Best reported score in this row) | – | 430.96 | – | 478.39 | 453.58 | – | – | – |
| BioMysteryBenchterminus v2 | 70.0% | – | 69.3% | – | – | – | – | – | – | 72.6% (Best reported score in this row) | 61.5% | 62.2% | 70.4% (Second-best reported score in this row) |
| Search and Tools |  |  |  |  |  |  |  |  |  |  |  |  |  |
| BrowseCompInkling 2 Small with compaction | 90.2% (Best reported score in this row) | 76.0% | 78.1% | 69.5% | 87.8% | 68.7% | 75.1% | – | 84.5% | 88.4% | 83.6% | 89.4% (Second-best reported score in this row) | 80.2% |
| DeepSearchQAInkling 2 Small with compaction | 90.9% (Second-best reported score in this row) | 84.3% | 86.0% | 82.7% | 84.5% | 80.4% | 84.7% | 90.0% | 84.4% | 95.0% (Best reported score in this row) | 83.5% | 90.1% | 83.2% |
| MCP AtlasFull / Public | 85.4/85.0% (Second-best reported score in this row) | –/82.0% | –/62.9% | –/80.4% | –/83.2% | 89.2/87.5% (Best reported score in this row) | 84.2/83.6% | –/84.4% | –/84.5% | 82.3/84.2% | –/77.8% | –/87.5% | –/81.0% |
| Knowledge Work |  |  |  |  |  |  |  |  |  |  |  |  |  |
| AA-Briefcase v1.1 | 1527 | 1451 | 1487 | 1476 | 1570 (Second-best reported score in this row) | 1279 | 1503 | 1512 | 1348 | 1497 | 1337 | 1203 | 1574 (Best reported score in this row) |
| GDPVal-AA v2.1 | 1630 | 1649 | 1614 | 1465 | 1746 (Best reported score in this row) | 1401 | 1657 | 1689 (Second-best reported score in this row) | 1384 | 1543 | 1444 | 1445 | 1631 |
| JobBench | 70.5% (Best reported score in this row) | 60.7% | 61.2% | 67.5% | 68.5% (Second-best reported score in this row) | 61.8% | 61.4% | 62.0% | 55.2% | 54.3% | 49.0% | 63.4% | 39.2% |
| DAYJOB (Healthcare) | 6.4% (Best reported score in this row) | 1.2% | – | – | 6.0% (Second-best reported score in this row) | – | 4.4% | – | 3.6% | 4.0% | 1.2% | 2.0% | – |
| Enterprise Operations |  |  |  |  |  |  |  |  |  |  |  |  |  |
| AutomationBench-AA v1.0.6 | 45.6% | 60.4% | 52.3% | 72.5% (Best reported score in this row) | 66.6% (Second-best reported score in this row) | 49.7% | 62.2% | 58.6% | 57.8% | 58.3% | 53.2% | 59.9% | 35.4% |
| EnterpriseOps-Gym | 46.8% (Second-best reported score in this row) | 33.2% | 29.0% | 48.9% (Best reported score in this row) | 43.6% | 43.1% | 36.4% | 38.0% | 44.9% | 45.3% | 24.9% | 45.7% | 25.1% |
| EnterpriseBench-CoreCraft | 71.8% (Best reported score in this row) | 46.7% | – | – | 58.5% (Second-best reported score in this row) | – | 51.8% | – | 47.7% | 43.1% | 43.1% | 58.5% (Second-best reported score in this row) | – |
| Finance Work |  |  |  |  |  |  |  |  |  |  |  |  |  |
| Finance Agent v2.1 | 67.5% (Best reported score in this row) | 60.6% | 63.3% | 58.8% | 62.9% | 60.7% | 65.0% | 64.3% | 61.7% | 60.0% | 56.8% | 65.8% (Second-best reported score in this row) | 65.4% |
| Excel Modeling Benchmark v1.1 | 69.5% | 66.3% | 68.9% | 67.8% | 69.2% | 62.0% | 69.1% | 67.4% | 68.7% | 71.3% (Second-best reported score in this row) | 70.0% | 72.9% (Best reported score in this row) | 65.5% |
| Tax Agent Bench | 65.8% | 62.0% | 59.9% | 62.7% | 74.9% (Best reported score in this row) | 52.1% | 73.1% (Second-best reported score in this row) | 64.9% | 67.5% | 68.7% | 58.9% | 66.8% | 62.2% |
| Apex-Accounting v1.1pass@1 / mean score | 14.5/62.0% | 12.3/53.7% | – | 13.5/60.0% | 12.5/59.3% | 13.8/58.0% | 13.9/59.5% | – | 12.0/56.5% | 9.5/56.7% | 10.8/53.9% | 15.5/58.8% (Best reported score in this row) | 15.3/62.2% (Second-best reported score in this row) |
| DAYJOB (Finance) | 14.3% (Best reported score in this row) | 7.3% | – | – | 12.5% (Second-best reported score in this row) | – | 11.0% | – | 4.8% | 5.5% | 6.8% | 4.5% | – |
| DeepFinance | 29.8% (Second-best reported score in this row) | – | – | – | – | – | 27.5% | – | 21.5% | – | 26.4% | – | 40.0% (Best reported score in this row) |
| Diligence BenchBest harness | 58.1% (Second-best reported score in this row) | 48.3% | – | 51.6% | 50.4% | 59.0% (Best reported score in this row) | 49.4% | 51.8% | 51.5% | 50.9% | 42.1% | 51.1% | 40.8% |
| AA-AnalystAgent | 51.2% (Second-best reported score in this row) | 25.0% | – | – | – | – | 38.8% | 37.5% | – | 38.8% | 43.8% | 46.3% | 53.8% (Best reported score in this row) |
| Tau 3 Banking v1.0.1 | 40.4% (Second-best reported score in this row) | 47.2% (Best reported score in this row) | – | – | – | – | – | – | – | 33.4% | – | – | – |
| ATLAS-Finance v1.1 | 6.0% (Best reported score in this row) | – | – | – | – | – | 3.0% | – | – | 3.7% (Second-best reported score in this row) | – | 3.7% (Second-best reported score in this row) | – |
| Legal Work |  |  |  |  |  |  |  |  |  |  |  |  |  |
| Harvey LAB v1.1 | 21.7% (Second-best reported score in this row) | 10.4% | 13.3% | 7.5% | 14.2% | 25.0% (Best reported score in this row) | 14.6% | 13.8% | 15.8% | 12.9% | 4.2% | 12.1% | 4.2% |
| Legal Research Bench | 46.6% | 45.2% | 38.0% | 34.1% | 45.2% | 38.5% | 49.0% (Best reported score in this row) | 47.1% (Second-best reported score in this row) | 35.6% | 46.2% | 30.3% | 38.9% | 43.8% |
| Visual Creation |  |  |  |  |  |  |  |  |  |  |  |  |  |
| DesignArena Web Apps | 1286 (Best reported score in this row) | – | – | – | – | – | 1271 | 1274 | – | 1280 (Second-best reported score in this row) | 1160 | 1224 | – |
| DesignArena Website | 1331 (Best reported score in this row) | 1266 | – | 1275 | 1296 | 1300 | 1285 | 1303 | 1253 | 1330 (Second-best reported score in this row) | 1265 | 1295 | 1257 |
| DesignArena Game Dev | 1424 (Best reported score in this row) | 1270 | – | 1307 | 1261 | 1254 | 1292 | 1309 | 1218 | 1351 (Second-best reported score in this row) | 1254 | 1298 | 1314 |
| DesignArena 3D Design | 1457 (Best reported score in this row) | 1300 | – | 1319 | 1344 | 1255 | 1304 | 1322 | 1239 | 1367 (Second-best reported score in this row) | 1199 | 1279 | 1329 |
| Code Arena (WebDev) | 1653 | 1607 | 1637 | 1582 | 1665 (Best reported score in this row) | – | 1623 | 1625 | 1544 | 1656 (Second-best reported score in this row) | 1579 | 1583 | 1589 |
| AA-3D | 1021 (Best reported score in this row) | 937 | – | – | – | – | 782 | 977 | – | 982 | – | 965 | 994 (Second-best reported score in this row) |
| Code2Video-AA | 1020 | 975 | 1052 (Second-best reported score in this row) | – | – | – | 1034 | 1052 (Second-best reported score in this row) | – | 1041 | 1030 | 899 | 1192 (Best reported score in this row) |
| BVB | 71.6% (Best reported score in this row) | 67.9% | – | – | 70.0% | – | – | – | – | – | 70.3% (Second-best reported score in this row) | 65.9% | 68.4% |
| Visual Perception |  |  |  |  |  |  |  |  |  |  |  |  |  |
| PerceptionBench | 64.7% (Second-best reported score in this row) | 49.6% | 39.1% | 47.1% | 47.7% | 44.1% | – | 38.9% | 46.5% | 58.5% | 56.2% | 68.7% (Best reported score in this row) | 60.5% |
| Humanity’s Sixth Sense | 33.3% (Second-best reported score in this row) | 26.8% | 25.4% | – | 30.1% | 29.6% | – | 25.2% | – | 25.5% | 21.0% | 41.6% (Best reported score in this row) | – |
| BabyVisionoriginal / with Python | 68.6/93.6% (Second-best reported score in this row) | 53.4/84.0% | 16.9/68.4% | 56.1/91.0% | 44.9/92.5% | 44.3/80.2% | – | 19.0/73.8% | 42.3/71.0% | 53.4/85.7% | 55.3/79.6% | 72.8/92.8% (Best reported score in this row) | 48.3/93.9% |
| ScreenSpot Prooriginal / with Python | 83.8/92.0% (Second-best reported score in this row) | –/85.3% | 35.4/72.5% | 48.5/76.8% | – | – | – | 56.6/79.7% | 42.3/43.4% | 73.8/86.1% | 81.1/87.7% | 65.7/93.7% | 85.4/88.7% (Best reported score in this row) |
| GroundingME | 62.1% (Best reported score in this row) | 21.3% | – | – | – | 51.1% (Second-best reported score in this row) | – | – | 38.9% | 42.2% | – | – | – |
| OmniDocBench v1.6 | 94.6% (Second-best reported score in this row) | 92.3% | 87.4% | 92.0% | 91.9% | 93.8% | – | 88.4% | 90.2% | 95.2% (Best reported score in this row) | 92.3% | 89.7% | 92.6% |
| SimpleVQAwith search and tools | 68.2% (Second-best reported score in this row) | 67.1% | – | – | – | 63.6% | – | – | 57.8% | 68.5% (Best reported score in this row) | – | – | – |
| Visual Reasoning |  |  |  |  |  |  |  |  |  |  |  |  |  |
| GDP.pdf | 25.2% (Best reported score in this row) | 15.4% | 9.0% | 20.8% | 17.2% | 18.6% | 11.2% | 19.2% | 17.6% | 22.0% | 22.8% (Second-best reported score in this row) | 21.0% | 20.8% |
| MMMU ProStandard 10 | 81.6% (Second-best reported score in this row) | 79.0% | 67.8% | 79.1% | 80.6% | 80.9% | – | 74.1% | 78.0% | 80.5% | 79.7% | 85.6% (Best reported score in this row) | 79.1% |
| Charxiv RQoriginal / with Python | 85.6/89.1% (Second-best reported score in this row) | 80.2/89.4% | 66.2/88.4% | 82.5/87.6% | 77.0/89.6% | 85.4/89.5% | – | 74.2/86.7% | 83.2/87.4% | 84.8/91.3% | 81.8/87.6% | 86.2/92.4% (Best reported score in this row) | 84.4/86.3% |
| Chartographyoriginal / with Python | 36.7/82.1% | 16.3/78.0% | 7.5/73.9% | 12.6%/– | 17.9/83.0% | 35.7/65.8% | – | 10.8/69.0% | 21.2/55.2% | 26.6/66.6% | 32.0/66.5% | 40.9/81.0% (Second-best reported score in this row) | 46.4/86.2% (Best reported score in this row) |
| MathVision | 92.2% | 82.9% | 51.6% | 89.1% | 89.4% | 87.7% | – | 68.3% | 88.9% | 93.1% (Second-best reported score in this row) | 90.0% | 93.3% (Best reported score in this row) | 92.1% |
| VIALSoriginal / with stirrup v0.2.1 | 34.6/57.8% (Best reported score in this row) | –/53.0% | – | –/49.7% | –/53.2% | – | – | –/58.4% | –/42.0% | 20.9%/– (Second-best reported score in this row) | –/44.1% | –/52.0% | –/60.5% |
| SAGE v1.1 | 88.6% | 89.3% (Second-best reported score in this row) | 83.3% | 85.3% | – | 85.8% | – | 84.4% | 74.2% | 86.8% | 83.4% | 88.2% | 89.8% (Best reported score in this row) |
| MuirBench | 79.2% (Second-best reported score in this row) | 71.6% | 63.2% | 69.4% | 75.7% | 74.5% | – | 72.6% | 67.5% | 77.0% | 73.9% | 80.1% (Best reported score in this row) | 74.0% |
| ZeroBench (main)original / with Python | 29.0/44.0% (Best reported score in this row) | 13.0/32.0% | 7.0%/– | 21.0%/– | 23.0%/– (Second-best reported score in this row) | 14.0/50.0% | – | 8.0%/– | 15.0/23.0% | 23.0/41.0% (Second-best reported score in this row) | 16.0%/– | 23.0%/– (Second-best reported score in this row) | 23.0%/– (Second-best reported score in this row) |
| STEM Reasoning |  |  |  |  |  |  |  |  |  |  |  |  |  |
| CritPt v2Full / Hard | 55.7/35.0% | –/24.5% | –/29.5% | 51.4/32.5% | 50.6/25.0% | 52.6/29.5% | 57.4/33.5% | 56.9/34.0% | 50.3/27.0% | –/40.5% | 65.1/42.5% (Best reported score in this row) | 65.1/46.5% (Best reported score in this row) | 64.0/43.5% |
| HLE-Diamond | 28.7% | 18.2% | 13.7% | 28.1% | 25.3% | 26.3% | 17.9% | 15.2% | 17.3% | 26.9% | 22.2% | 33.1% (Best reported score in this row) | 29.6% (Second-best reported score in this row) |
| HLE (text only) | 44.1% | 39.9% | 35.1% | 48.0% (Second-best reported score in this row) | 43.3% | 43.9% | 42.3% | 49.4% (Best reported score in this row) | 36.4% | 46.9% | 38.5% | 47.8% | 44.4% |
| GPQA Diamond | 93.0% | 91.2% | 88.7% | 86.7% | 92.2% | 89.0% | 91.7% | 89.0% | 89.7% | 93.5% (Second-best reported score in this row) | 90.5% | 95.3% (Best reported score in this row) | 89.8% |
| Apex 2025 | 63.5% | 26.0% | 12.0% | 28.1% | 68.8% (Best reported score in this row) | 32.3% | 29.2% | 9.9% | 45.8% | 65.6% (Second-best reported score in this row) | 45.8% | 57.3% | 17.2% |
| HMMT Feb 2026 | 95.8% | 93.2% | 87.1% | 92.4% | 97.7% (Best reported score in this row) | 93.9% | 95.5% | 88.6% | 95.5% | 97.0% (Second-best reported score in this row) | 90.9% | 95.5% | 50.0% |
| ArxivMath10/2026 | 27.7% (Second-best reported score in this row) | – | – | – | 38.3% (Best reported score in this row) | 19.2% | – | – | 18.1% | – | – | – | – |
| BrokenArxiv10/2026 | 63.3% (Best reported score in this row) | – | – | – | 49.7% (Second-best reported score in this row) | 33.3% | – | – | 30.6% | – | – | – | – |
| ARC-AGI-1 | 96.5% (Second-best reported score in this row) | 91.0% | – | – | 93.5% | 89.0% | – | – | – | 94.5% | 86.7% | 98.5% (Best reported score in this row) | 89.0% |
| ARC-AGI-2 | 81.7% (Second-best reported score in this row) | 65.8% | – | – | 78.1% | 57.5% | – | – | – | 60.4% | 59.3% | 89.2% (Best reported score in this row) | 61.3% |
| MysteryMechanism | 25.2% | – | 21.6% | – | – | – | 23.0% | 15.3% | – | – | 19.4% | 36.5% (Best reported score in this row) | 35.1% (Second-best reported score in this row) |
| Long Context |  |  |  |  |  |  |  |  |  |  |  |  |  |
| AA-LCR v1.1 | 83.3% | 80.0% | 74.3% | 79.7% | 82.7% | 78.3% | 79.7% | 86.3% (Second-best reported score in this row) | 78.0% | 88.7% (Best reported score in this row) | 83.3% | 81.3% | 82.7% |
| LongBench v2 | 70.6% (Second-best reported score in this row) | 68.2% | 60.7% | 60.9% | 69.0% | 68.7% | 68.4% | 68.0% | 65.7% | 69.0% | 65.0% | 73.7% (Best reported score in this row) | 65.3% |
| Factuality |  |  |  |  |  |  |  |  |  |  |  |  |  |
| AA-Omniscience v1.1score | 16.1 | 0.7 | -10.8 | -6.4 | 0.9 | 25.8 (Best reported score in this row) | 6.0 | -4.5 | 5.1 | – | -8.3 | 23.3 (Second-best reported score in this row) | 12.3 |
| AA-Omniscience v1.1non-hallucination rate | 73.3% (Best reported score in this row) | 61.7% | 47.8% | 6.3% | 35.9% | 72.0% (Second-best reported score in this row) | 61.1% | 39.8% | 71.8% | – | 20.7% | 44.9% | 65.8% |
| Instruction Following |  |  |  |  |  |  |  |  |  |  |  |  |  |
| ComplexConstraints v1.0 | 44.5% | 27.6% | 22.1% | 23.2% | 50.4% (Best reported score in this row) | 41.3% | 30.5% | 32.3% | 36.5% | 37.9% | 27.3% | 48.4% (Second-best reported score in this row) | 36.8% |
| Multilinguality |  |  |  |  |  |  |  |  |  |  |  |  |  |
| Global-MMLU-Lite | 91.0% | 89.0% | 83.9% | 78.5% | 90.4% | 91.2% (Second-best reported score in this row) | 89.8% | 88.0% | 88.0% | 90.6% | 90.2% | 93.2% (Best reported score in this row) | 87.8% |
| MDPBench | 81.4% | 70.9% | 54.7% | 69.2% | 68.8% | 82.0% (Second-best reported score in this row) | – | 67.0% | 64.3% | 83.6% (Best reported score in this row) | 71.2% | 81.5% | 78.4% |
| Safety |  |  |  |  |  |  |  |  |  |  |  |  |  |
| HoneyBench v0.2 (↓) | 52.6% (Best reported score in this row) | – | – | – | 76.4% (Second-best reported score in this row) | 85.6% | 76.8% | – | 83.6% | 78.5% | – | 80.5% | – |
| DistressBench | 82.0% (Second-best reported score in this row) | 72.7% | – | – | – | 87.7% (Best reported score in this row) | 72.5% | – | 77.0% | 73.5% | 66.7% | 70.7% | 71.7% |
| FORTRESS (adversarial) | 83.8% | 73.3% | 64.7%* | 47.9% | 72.8% | 92.4% (Best reported score in this row) | 71.8% | 69.9%* | 76.0% | 73.4% | 89.2% | 79.5%* | 90.5% (Second-best reported score in this row) |
| FORTRESS (benign) | 79.7% | 96.4% | 97.4%* (Best reported score in this row) | 96.8% | 93.1% | 84.0% | 96.8% | 97.1%* | 91.8% | 97.3% (Second-best reported score in this row) | 92.4% | 94.8%* | 91.0% |
| StrongREJECT (none) | 97.1% | 97.8% | 97.8% | 97.4% | 99.4% (Best reported score in this row) | 97.1% | 97.5% | 99.0% | 99.0% | 99.0% | 99.4% (Best reported score in this row) | 93.0% | 97.8% |
| Cybersecurity |  |  |  |  |  |  |  |  |  |  |  |  |  |
| CyberBench v1.1 | 72.7% | 67.9% | 75.4% | – | – | – | 72.1% | 72.9% | – | 75.2% | 76.3% (Best reported score in this row) | 43.8% | 75.5% (Second-best reported score in this row) |
| PatchEval | 79.6% | 74.8% | 77.0% | 76.5% | 80.4% (Second-best reported score in this row) | 77.8% | 69.6% | 80.0% | 76.1% | 77.0% | 76.5% | 77.8% | 81.7% (Best reported score in this row) |
* On public split only.
For ScreenSpot Pro evaluation, images whose long side exceed 4,096 pixels are downsampled to that limit while preserving the aspect ratio; the ground truth answers were rescaled accordingly. Smaller images remain unchanged.
6. Safety
We conducted safety evaluations ahead of release, spanning both everyday human-AI interaction and dangerous-capability testing. Because Inkling 2 Small is multimodal, we paid attention to whether safety behavior held consistently across text and image inputs. We applied mitigations to reduce risks before release.
For everyday interaction, we evaluated sycophancy, harmful manipulation, and psychological-harm patterns like parasocial dependency and validation of delusional reasoning, including through multi-turn, open-ended external red-teaming designed to surface issues that only emerge over longer conversations. We also assessed whether the model refuses genuinely harmful requests without over-refusing benign ones. For CBRN and cyber, we assessed knowledge and procedural uplift through internal evaluations, external testing, and refusal-suppressed variants intended to estimate latent capability with safeguards removed. For loss of control, we evaluated agentic capability, strategic deception, and sabotage potential, benchmarked against public frontier models, and found the model materially below frontier capabilities.
Across all areas, we concluded that Inkling 2 Small did not present risk of material uplift beyond what’s already available in the open-weight ecosystem.
7. Bias, risks and limitations
Inkling 2 Small may exhibit general limitations common to foundation models, including hallucination (generating plausible but factually incorrect or unsupported content), occasional failures to follow instructions precisely, and degraded performance in long multi-turn conversations. As with other large-scale models trained on web-derived and synthetic data, Inkling 2 Small may reflect biases present in its training data, including demographic, cultural, or linguistic biases, and may perform unevenly across languages, dialects, or subject domains that were less represented during training.
Inkling 2 Small’s knowledge is limited to information available as of its training cutoff, and it may not reflect events, developments, or changes that occurred afterward.
We recommend that downstream developers and deployers apply appropriate human oversight and review for outputs used in high-stakes or safety-critical contexts, rather than relying on Inkling 2 Small’s outputs without verification.
- Conduct their own evaluation of Inkling 2 Small’s performance, safety, and fairness for their specific use case, language, and population prior to deployment, particularly for applications involving vulnerable groups.
- Implement additional safeguards – such as content filtering, rate limiting, and monitoring – at the application layer, especially for open deployment contexts where Inkling 2 Small’s built-in mitigations may not be sufficient on their own.
- Avoid deploying Inkling 2 Small in domains such as medical, legal, or safety-critical decision-making without additional fine-tuning, domain-specific validation, and human oversight