Inkling 2 Small — EXL3 3.17-bit experts
A 250.6 GB quantization of thinkingmachines/Inkling-2-Small, built to serve on 3x NVIDIA RTX PRO 6000 Blackwell (96 GB) at tensor parallel 3 with a 1M-token context.
Quantization
- Routed experts: EXL3 (trellis quantization,
mul1codebook, output scales). Layers 0–39 are 3 bpw and layers 40–47 are 4 bpw, for a 3.17 bpw expert average. The source is the NVFP4 release: its experts were dequantized, then requantized. - Everything else (attention, router, embeddings, vision encoder, DSpark/MTP drafter) is unchanged from the parent checkpoint.
- Calibration: 512k tokens (250 rows x 2,048) from exllamav3's default calibration mix of web, code, multilingual, technical, wiki and random-token text, propagated layer by layer through the parent model. Each layer's gate/up projections share one input Hessian. Each expert's down projection gets its own Hessian, computed over all calibration tokens.
Accuracy vs. the NVFP4 checkpoint
Prompt log-probabilities on a held-out panel of 16 windows x 1,024 tokens (16,368 scored positions): 8 technical-prose pages from the vLLM docs and 8 CPython standard-library modules, none drawn from the calibration data. Both models run through the same layer-by-layer reference implementation, with the NVFP4 experts dequantized to BF16, so the comparison isolates the expert quantization. KL is over the full vocabulary, KL(NVFP4 ‖ EXL3).
| Panel | NLL, NVFP4 | NLL, this model | Δ NLL | Mean KL | Top-1 agreement |
|---|---|---|---|---|---|
| Prose (8,184 positions) | 1.0366 | 1.0400 | +0.0034 | 0.0100 | 96.80% |
| Code (8,184 positions) | 0.3493 | 0.3610 | +0.0118 | 0.0172 | 97.47% |
| All (16,368 positions) | 0.6929 | 0.7005 | +0.0076 | 0.0136 | 97.13% |
A +0.0076 NLL change is a 0.76% increase in perplexity.
Compatibility
This is not a drop-in checkpoint. It does not load in stock vLLM, SGLang, Transformers, or ExLlamaV3's standard server. It runs on a patched vLLM with ExLlamaV3 kernels.
The checkpoint keeps the original 48 query / 8 KV head layout, so TP 1, 2, 4 and 8 divide it evenly; TP 3 works through the runtime, which pads attention to 54 / 9 heads at load time (the added heads are inert, so outputs are unchanged). The weights are not tied to one GPU generation: the EXL3 kernels run on NVIDIA Ampere (SM80) and newer. Only the configuration below has been tested.
- Runtime source and documentation: meldavy/inkling2-small-sm120-tp3: patches, build steps, launch options and the Docker setup.
- Docker image:
ghcr.io/meldavy/inkling2-small-sm120-tp3:latest(linux/amd64,linux/arm64). The host needs an NVIDIA driver with CUDA 13.0 support (580 or newer) and the NVIDIA Container Toolkit.
Sample docker-compose.yml for 3x RTX PRO 6000 (TP3, FP8 KV cache, 1M context, DSpark speculative decoding):
services:
inkling2:
image: ghcr.io/meldavy/inkling2-small-sm120-tp3:latest
ipc: host
shm_size: 32g
ulimits:
memlock: -1
ports:
- "8300:8300"
volumes:
- ${MODEL_DIR:?set MODEL_DIR}:/model:ro
- ./cache:/root/.cache
environment:
KV_DTYPE: fp8 # fp8 | auto (BF16)
SPEC: "1" # DSpark speculative decoding; 0 to disable
SPEC_K: "3" # drafts verified per step with one request
SPEC_K_BATCH: "2" # drafts per request with two or more requests
MAXLEN: "1000000" # context window
UTIL: "0.98" # GPU memory fraction; 0.98 recommended for a 1M-token context
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: 3
capabilities: [gpu]
restart: unless-stopped
hf download mel-bw/Inkling-2-Small-EXL3-3.17bpw --local-dir /path/to/Inkling-2-Small-EXL3-3.17bpw
MODEL_DIR=/path/to/Inkling-2-Small-EXL3-3.17bpw docker compose up -d
The server is OpenAI-compatible on port 8300, served as Inkling-2-Small. Work is tracked in the Local Inference Lab Discord.
Performance
Measured on 3x RTX PRO 6000 Blackwell Max-Q (250 W power limit each), TP3, FP8 KV cache, 1,000,000-token context, prefix caching on, single stream, no speculative decoding:
| Throughput | |
|---|---|
| Prefill, 8k / 32k prompt | ~5,400 / ~5,500 tok/s |
| Decode | ~105 tok/s |
With a BF16 KV cache, the same setup fits an 850,000-token context at the same speed. Other configurations, DSpark speculative decoding and quality benchmarks are in the runtime repo.
Credits
- Thinking Machines for Inkling 2 Small (Apache-2.0).
- turboderp for EXL3 / exllamav3.
- diffbot for the decode-once EXL3 MoE prefill kernel (DeepSeek-V4.1-Flash recipe).
- vLLM: Inkling 2 support (#67325, WoosukKwon); Triton attention fallbacks for Inkling on SM12x (#51560, m4xkushnir) and SM8x (#55078, TheVerdman).
- Local Inference Lab: the KK vLLM fork's TP3 head padding, followed by the runtime.
- Evaluation text: the vLLM documentation and the CPython standard library (Python Software Foundation).
- Runtime credits (vLLM, FlashInfer, NCCL, benchmarks and community findings): see the runtime repo.
- The per-layer EXL3 expert layout follows the DeepSeek-V4.1-Flash EXL3 quants by LibertAIDAI and Mia-AiLab.
The original model card follows.
Inkling 2 Small
NVFP4 | MXFP8 | Playground | Tinker Cookbook | Acceptable Use
1. General Information
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, with general multilingual capabilities across other languages.
2. Getting Started
Try Inkling 2 Small on the Tinker Playground or access via API using the Tinker Cookbook. To fine-tune Inkling 2 Small with Tinker, also install the tml-renderers package.
Inkling 2 Small supports local inference using:
Inkling 2 Small can also be post-trained with:
API access is also available through third party inference providers, including:
- NVIDIA
- Fireworks
- Together
- Baseten
- Nebius
- Modal
- Cloudflare
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.
3. 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, 17B active
Numerics support
MXFP8 and NVFP4
Input modalities
Inkling 2 Small accepts text input in UTF-8 encoding and image input in any pixel-based format (with each dimension ideally between 40px and 4096px for optimal performance). It supports a context window of up to 1M tokens.
Output modalities
Inkling 2 Small generates output as UTF-8 encoded text.
4. Training
Training data includes a broad variety of content types, including text, images, video. 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.
The training data curation process 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 | 24.9 | 31.7 | 29.9 | 26.1 | – | 27.9 | 30.8 | 36.5 |
| Vals Index (v2.2) | 56.1% | 50.9% | 54.8% | 53.8% | 61.3% | 48.6% | 59.7% | 57.8% | 56.5% | 53.0% | 52.3% | 56.0% | 58.7% |
| Coding | |||||||||||||
| FrontierCode v2.0 stirrup v0.2.1 | 25.4% | 17.6% | 21.6% | 20.8% | 28.5% | 7.6% | 21.6% | 24.0% | 15.0% | 17.3% | 25.2% | 18.7% | 34.9% |
| DeepSWE v1.1 mini-swe-agent v2.4.6 | 72.1% | 63.4% | 67.9% | 73.7% | 77.0% | 60.0% | 69.0% | 71.9% | 61.7% | 68.5% | 66.6% | 73.8% | 59.9% |
| Terminal Bench Science v0.1 mini-swe-agent v2.4.6 | 5.7% | 4.8% | 4.3% | 3.3% | 12.9% | 1.9% | 9.5% | 5.7% | 4.3% | 8.1% | 8.6% | 10.0% | 20.0% |
| Vibe Code Bench v1.1 openhands (vals) | 81.6% | 30.8% | 79.0% | 73.9% | 87.7% | 76.2% | 78.1% | 85.2% | 81.6% | 85.0% | 81.7% | 78.7% | 90.4% |
| SWE-Atlas v1.1 mini-swe-agent v2.4.6 | 61.9% | – | – | – | 62.1% | 42.3% | – | – | – | 61.0% | 50.8% | 57.3% | 59.4% |
| SWEBench Pro v2 (Hard) mini-swe-agent v2.4.6 | 86.3% | 82.8% | 89.7% | 84.3% | 89.7% | 82.4% | 84.3% | 89.2% | 83.8% | 88.2% | 88.7% | 58.8% | 52.9% |
| Sudo L7 | 14.1% | – | – | – | – | – | 9.4% | – | 13.9% | 12.2% | – | 10.6% | – |
| Autoresearch | |||||||||||||
| PostTrainBench v1.2 Inkling 2 Small in Claude Code | 31.9% | 36.8% | – | – | 30.5% | – | 34.5% | – | – | 31.5% | – | – | – |
| PMPP-Hard (AUC, 0–10h) pi v0.83.0 | 494.27 | – | – | – | 497.36 | – | 430.96 | – | 478.39 | 453.58 | – | – | – |
| BioMysteryBench terminus v2 | 70.0% | – | 69.3% | – | – | – | – | – | – | 72.6% | 61.5% | 62.2% | 70.4% |
| Search and Tools | |||||||||||||
| BrowseComp Inkling 2 Small with compaction | 90.2% | 76.0% | 78.1% | 69.5% | 87.8% | 68.7% | 75.1% | – | 84.5% | 88.4% | 83.6% | 89.4% | 80.2% |
| DeepSearchQA Inkling 2 Small with compaction | 90.9% | 84.3% | 86.0% | 82.7% | 84.5% | 80.4% | 84.7% | 90.0% | 84.4% | 95.0% | 83.5% | 90.1% | 83.2% |
| MCP Atlas Full / Public | 85.4/85.0% | –/82.0% | –/62.9% | –/80.4% | –/83.2% | 89.2/87.5% | 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 | 1279 | 1503 | 1512 | 1348 | 1497 | 1337 | 1203 | 1574 |
| GDPVal-AA v2.1 | 1630 | 1649 | 1614 | 1465 | 1746 | 1401 | 1657 | 1689 | 1384 | 1543 | 1444 | 1445 | 1631 |
| JobBench | 70.5% | 60.7% | 61.2% | 67.5% | 68.5% | 61.8% | 61.4% | 62.0% | 55.2% | 54.3% | 49.0% | 63.4% | 39.2% |
| DAYJOB (Healthcare) | 6.4% | 1.2% | – | – | 6.0% | – | 4.4% | – | 3.6% | 4.0% | 1.2% | 2.0% | – |
| Enterprise Operations | |||||||||||||
| EnterpriseOps-Gym | 46.8% | 33.2% | 29.0% | 48.9% | 43.6% | 43.1% | 36.4% | 38.0% | 44.9% | 45.3% | 24.9% | 45.7% | 25.1% |
| EnterpriseBench-CoreCraft | 71.8% | 46.7% | – | – | 58.5% | – | 51.8% | – | 47.7% | 43.1% | 43.1% | 58.5% | – |
| Finance Work | |||||||||||||
| Finance Agent v2.1 | 67.5% | 60.6% | 63.3% | 58.8% | 62.9% | 60.7% | 65.0% | 64.3% | 61.7% | 60.0% | 56.8% | 65.8% | 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% | 70.0% | 72.9% | 65.5% |
| Tax Agent Bench | 65.8% | 62.0% | 59.9% | 62.7% | 74.9% | 52.1% | 73.1% | 64.9% | 67.5% | 68.7% | 58.9% | 66.8% | 62.2% |
| Apex-Accounting v1.1 pass@1 | 14.5% | 12.3% | – | 13.5% | 12.5% | 13.8% | 13.9% | – | 12.0% | 9.5% | 10.8% | 15.5% | 15.3% |
| DAYJOB (Finance) | 14.3% | 7.3% | – | – | 12.5% | – | 11.0% | – | 4.8% | 5.5% | 6.8% | 4.5% | – |
| DeepFinance | 29.8% | – | – | – | – | – | 27.5% | – | 21.5% | – | 26.4% | – | 40.0% |
| Diligence Bench Best harness | 58.1% | 48.3% | – | 51.6% | 50.4% | 59.0% | 49.4% | 51.8% | 51.5% | 50.9% | 42.1% | 51.1% | 40.8% |
| AA-AnalystAgent | 51.2% | 25.0% | – | – | – | – | 38.8% | 37.5% | – | 38.8% | 43.8% | 46.3% | 53.8% |
| Tau 3 Banking v1.0.1 | 40.4% | 47.2% | – | – | – | – | – | – | – | 33.4% | – | – | – |
| ATLAS-Finance v1.1 | 6.0% | – | – | – | – | – | 3.0% | – | – | 3.7% | – | 3.7% | – |
| Legal Work | |||||||||||||
| Harvey LAB v1.1 | 21.7% | 10.4% | 13.3% | 7.5% | 14.2% | 25.0% | 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% | 47.1% | 35.6% | 46.2% | 30.3% | 38.9% | 43.8% |
| Visual Creation | |||||||||||||
| DesignArena Web Apps | 1286 | – | – | – | – | – | 1271 | 1274 | – | 1280 | 1160 | 1224 | – |
| DesignArena Website | 1331 | 1266 | – | 1275 | 1296 | 1300 | 1285 | 1303 | 1253 | 1330 | 1265 | 1295 | 1257 |
| DesignArena Game Dev | 1424 | 1270 | – | 1307 | 1261 | 1254 | 1292 | 1309 | 1218 | 1351 | 1254 | 1298 | 1314 |
| DesignArena 3D Design | 1457 | 1300 | – | 1319 | 1344 | 1255 | 1304 | 1322 | 1239 | 1367 | 1199 | 1279 | 1329 |
| Code Arena (WebDev) | 1653 | 1607 | 1637 | 1582 | 1665 | – | 1623 | 1625 | 1544 | 1656 | 1579 | 1583 | 1589 |
| AA-3D | 1021 | 937 | – | – | – | – | 782 | 977 | – | 982 | – | 965 | 994 |
| Code2Video-AA | 1020 | 975 | 1052 | – | – | – | 1034 | 1052 | – | 1041 | 1030 | 899 | 1192 |
| BVB | 71.6% | 67.9% | – | – | 70.0% | – | – | – | – | – | 70.3% | 65.9% | 68.4% |
| Visual Perception | |||||||||||||
| PerceptionBench | 64.7% | 49.6% | 39.1% | 47.1% | 47.7% | 44.1% | – | 38.9% | 46.5% | 58.5% | 56.2% | 68.7% | 60.5% |
| Humanity’s Sixth Sense | 33.3% | 26.8% | 25.4% | – | 30.1% | 29.6% | – | 25.2% | – | 25.5% | 21.0% | 41.6% | – |
| BabyVision original / with Python | 68.6/93.6% | 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% | 48.3/93.9% |
| ScreenSpot Pro original / with Python | 83.8/92.0% | –/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% |
| OmniDocBench v1.6 | 94.6% | 92.3% | 87.4% | 92.0% | 91.9% | 93.8% | – | 88.4% | 90.2% | 95.2% | 92.3% | 89.7% | 92.6% |
| SimpleVQA with search and tools | 68.2% | 67.1% | – | – | – | 63.6% | – | – | 57.8% | 68.5% | – | – | – |
| Visual Reasoning | |||||||||||||
| GDP.pdf | 25.2% | 15.4% | 9.0% | 20.8% | 17.2% | 18.6% | 11.2% | 19.2% | 17.6% | 22.0% | 22.8% | 21.0% | 20.8% |
| MMMU Pro Standard 10 | 81.6% | 79.0% | 67.8% | 79.1% | 80.6% | 80.9% | – | 74.1% | 78.0% | 80.5% | 79.7% | 85.6% | 79.1% |
| Charxiv RQ original / with Python | 85.6/89.1% | 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% | 84.4/86.3% |
| Chartography original / 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% | 46.4/86.2% |
| VIALS original / with stirrup v0.2.1 | 34.6/57.8% | –/53.0% | – | –/49.7% | –/53.2% | – | – | –/58.4% | –/42.0% | 20.9%/– | –/44.1% | –/52.0% | –/60.5% |
| SAGE v1.1 | 88.6% | 89.3% | 83.3% | 85.3% | – | 85.8% | – | 84.4% | 74.2% | 86.8% | 83.4% | 88.2% | 89.8% |
| MuirBench | 79.2% | 71.6% | 63.2% | 69.4% | 75.7% | 74.5% | – | 72.6% | 67.5% | 77.0% | 73.9% | 80.1% | 74.0% |
| ZeroBench (main) original / with Python | 29.0/44.0% | 13.0/32.0% | 7.0%/– | 21.0%/– | 23.0%/– | 14.0/50.0% | – | 8.0%/– | 15.0/23.0% | 23.0/41.0% | 16.0%/– | 23.0%/– | 23.0%/– |
| STEM Reasoning | |||||||||||||
| CritPt v2 Full / 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% | 65.1/46.5% | 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% | 29.6% |
| HLE (text only) | 44.1% | 39.9% | 35.1% | 48.0% | 43.3% | 43.9% | 42.3% | 49.4% | 36.4% | 46.9% | 38.5% | 47.8% | 44.4% |
| ArxivMath 10/2026 | 27.7% | – | – | – | 38.3% | 19.2% | – | – | 18.1% | – | – | – | – |
| BrokenArxiv 10/2026 | 63.3% | – | – | – | 49.7% | 33.3% | – | – | 30.6% | – | – | – | – |
| ARC-AGI-1 | 96.5% | 91.0% | – | – | 93.5% | 89.0% | – | – | – | 94.5% | 86.7% | 98.5% | 89.0% |
| ARC-AGI-2 | 81.7% | 65.8% | – | – | 78.1% | 57.5% | – | – | – | 60.4% | 59.3% | 89.2% | 61.3% |
| MysteryMechanism | 25.2% | – | 21.6% | – | – | – | 23.0% | 15.3% | – | – | 19.4% | 36.5% | 35.1% |
| Long Context | |||||||||||||
| AA-LCR v1.1 | 83.3% | 80.0% | 74.3% | 79.7% | 82.7% | 78.3% | 79.7% | 86.3% | 78.0% | 88.7% | 83.3% | 81.3% | 82.7% |
| LongBench v2 | 70.6% | 68.2% | 60.7% | 60.9% | 69.0% | 68.7% | 68.4% | 68.0% | 65.7% | 69.0% | 65.0% | 73.7% | 65.3% |
| Factuality | |||||||||||||
| AA-Omniscience v1.1 score | 16.1 | 0.7 | -10.8 | -6.4 | 0.9 | 25.8 | 6.0 | -4.5 | 5.1 | – | -8.3 | 23.3 | 12.3 |
| AA-Omniscience v1.1 non-hallucination rate | 73.3% | 61.7% | 47.8% | 6.3% | 35.9% | 72.0% | 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% | 41.3% | 30.5% | 32.3% | 36.5% | 37.9% | 27.3% | 48.4% | 36.8% |
| Multilinguality | |||||||||||||
| Global-MMLU-Lite | 91.0% | 89.0% | 83.9% | 78.5% | 90.4% | 91.2% | 89.8% | 88.0% | 88.0% | 90.6% | 90.2% | 93.2% | 87.8% |
| MDPBench | 81.4% | 70.9% | 54.7% | 69.2% | 68.8% | 82.0% | – | 67.0% | 64.3% | 83.6% | 71.2% | 81.5% | 78.4% |
| Safety | |||||||||||||
| HoneyBench v0.2 (↓) | 52.6% | – | – | – | 76.4% | 85.6% | 76.8% | – | 83.6% | 78.5% | – | 80.5% | – |
| DistressBench | 82.0% | 72.7% | – | – | – | 87.7% | 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% | 71.8% | 69.9%* | 76.0% | 73.4% | 89.2% | 79.5%* | 90.5% |
| FORTRESS (benign) | 79.7% | 96.4% | 97.4%* | 96.8% | 93.1% | 84.0% | 96.8% | 97.1%* | 91.8% | 97.3% | 92.4% | 94.8%* | 91.0% |
| StrongREJECT (none) | 97.1% | 97.8% | 97.8% | 97.4% | 99.4% | 97.1% | 97.5% | 99.0% | 99.0% | 99.0% | 99.4% | 93.0% | 97.8% |
| Cybersecurity | |||||||||||||
| CyberBench v1.1 | 72.7% | 67.9% | 75.4% | – | – | – | 72.1% | 72.9% | – | 75.2% | 76.3% | 43.8% | 75.5% |
| PatchEval | 79.6% | 74.8% | 77.0% | 76.5% | 80.4% | 77.8% | 69.6% | 80.0% | 76.1% | 77.0% | 76.5% | 77.8% | 81.7% |
Best reported score in each row (lowest where lower is better); equal displayed scores tie Second-best reported score in each row; none when first place is tied
* 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.
8. Legal
- Downloads last month
- 193
Model tree for mel-bw/Inkling-2-Small-EXL3-3.17bpw
Base model
thinkingmachines/Inkling-2-Small