Inkling vs Qwen3-VL-2B-Instruct

Inkling

Thinking Machines' large multimodal generalist, stronger than Inkling-Small on factual recall by its own numbers. It needs about 710 GB of datacentre GPU memory even in NVFP4, so most agents will use a host such as Together AI or Tinker. The main risks are a use policy that binds on download and can change, and no outside safety results.

Qwen3-VL-2B-Instruct

The smallest Qwen3-VL cousin, for reading screenshots, documents, charts and video on a laptop or edge box, running in about 2 GB from Alibaba's Q4_K_M GGUF in every major runtime under plain Apache 2.0. Qwen hasn't answered repository questions or published safety work, so keep a human or a check in the loop before it drives a GUI.

Inkling

1. Unless you have a multi-GPU Blackwell or H200 node, call it through Together AI, Fireworks, DeepInfra or Tinker rather than starting it
2. On 8 B200 GPUs, SGLang's recipe is sglang serve --trust-remote-code --model-path thinkingmachines/Inkling-NVFP4 --tp 8 --quantization modelopt_fp4 --enable-multimodal --reasoning-parser inkling --tool-call-parser inkling
3. Use the inkling tool parser, not Hermes, and the inkling reasoning parser
4. Pass reasoning_effort as none, minimal, low, medium, high (the default) or max, or a number from 0 to 0.99. The repository's template has no xhigh preset and raises an error for it, though Thinking Machines' docs list xhigh
5. Tool call arguments must reach the template as an object, not a JSON string, or it raises an error
6. Audio should be 16 kHz WAV, ideally under 20 minutes, and images 40 to 4096 px a side, per the card
7. Read the Model Acceptable Use Policy before deploying, it binds on download and bars some uses that are legal

Qwen3-VL-2B-Instruct

1. Start it with llama-server -hf Qwen/Qwen3-VL-2B-Instruct-GGUF:Q4_K_M --jinja -c 32768, or ollama run qwen3-vl:2b-instruct
2. In Ollama the plain qwen3-vl:2b tag is the Thinking edition, so name 2b-instruct
3. With the GGUF, load the mmproj vision file too, or images won't be read
4. Parse tool calls as JSON inside <tool_call> tags, or use SGLang's --tool-call-parser qwen
5. Use the card's sampling, temperature=0.7, top_p=0.8, top_k=20, presence_penalty=1.5 for vision and temperature=1.0, top_p=1.0, top_k=40, presence_penalty=2.0 for text
6. Validate JSON output yourself, since structured output isn't documented
7. vLLM needs 0.11.0 or later and Transformers 4.57.0 or later