Nanbeige4.2-3B vs Qwen3.5-0.8B

Nanbeige4.2-3B 72

Model card 17, Intended use, limitations, the architecture, the parameters (3B outside the embeddings), the context and the languages (in the metadata) are on the card. Training data 5, A general description of how the fine-tuning and reinforcement learning data was built and filtered, with no sources, sizes or cut-off. Last change 15, the last commit, on 11 September 2026, updated modeling_nanbeige.py, 60 days before this check and within 90. Responsiveness 20, Of the 34 most recent discussions that aren't pull requests (21 July to 10 September 2026), 20 have a reply from a Nanbeige member, which is most. Version policy 5, Nanbeige's earlier models (Nanbeige4-3B, 4.1-3B) stay in their own repositories, as the Base, FP8, 8-bit and DSpark builds of this one do. File health 10, Config, tokenizer config, template and card agree on the architecture and the 262,144 tokens. 17 + 5 + 15 + 20 + 5 + 10 = 72.

Qwen3.5-0.8B 37

The card covers six of the seven items, missing only a cut-off (17), because it states the intended uses and warns about thinking loops. Training data is a general description without a cut-off (5). The files last changed on 2 March 2026, 248 days before this check, within a year (5). No Qwen member has answered any of the 15 discussions that aren't pull requests (0). Qwen3-0.6B stays in its own repository without a dated note (5), and a template fix and the missing generation_config.json wait unanswered (5). 17+5+5+0+5+5 = 37.

Nanbeige4.2-3B 85

Chat template 20, A Jinja chat template in tokenizer_config.json (7,961 characters), with enable_thinking, preserve_thinking, tools and tool_call_format switches. Tool-calling format 25, the card says passing tools switches on the tool template and recommends tool_call_format xml, with json for compatibility, and its SGLang and vLLM commands name the nanbeige tool-call parser, which released SGLang 0.5.23 has. Trained tool calling 20, the card describes fine-tuning on tool-use environments and agent scaffolds and reinforcement learning on agent tasks, and the template tells the model it may call one or more functions and renders every call in a turn. Structured output 0, Neither the card nor the files mention JSON or JSON Schema output, beyond json as a tool-call format. Reasoning control 10, the card documents enable_thinking, on by default and False per request, and preserve_thinking for earlier turns, and names the nanbeige reasoning parser, which returns the reasoning apart from the answer. Context 10, 262,144 tokens, which the card says the model supports and config.json gives as max_position_embeddings. 20 + 25 + 20 + 0 + 10 + 10 = 85.

Qwen3.5-0.8B 60

The template ships in the repository (20) and the model was trained for tool use with several calls per turn in the template (20). Tool-calling format gets 10, because the card names the qwen3_coder parser but leaves the format to the template. Reasoning control scores 0, since the card shows how to turn thinking on but not where the reasoning comes back. 262,144 tokens is native (10). Structured output scores 0, as nothing documents JSON output for the model. 20+10+20+0+0+10 = 60.

Nanbeige4.2-3B 85

Memory 30, the smallest build from Nanbeige is the FP8 repository (the 8-bit GPTQ one is the same size), 5.19 GB, and 5.19 GB × 1.2 = 6.2 GB, within 16 GB. bartowski's GGUF isn't read for this line. Runtimes 15, Released SGLang 0.5.23 has the model (srt/models/nanbeige.py) with its parsers, llama.cpp b12242 has the nanbeige architecture, and vLLM 0.33.0 names NanbeigeForCausalLM in its registry, served through its Transformers backend. Builds 15, Quantised builds from Nanbeige at two precisions, FP8 and 8-bit GPTQ, each in its own repository. Terms 15, Apache 2.0 in the card's metadata, an OSI licence, with no use policy or other terms on the card. Download 10, the Hub API reports gated false for the model and its FP8 and 8-bit repositories on 10 November 2026, so the files download with no account and nothing to accept. Hosted 0, Hugging Face maps no inference provider to it, and no Nanbeige or BOSS Zhipin API for it was found. 30 + 15 + 15 + 15 + 10 + 0 = 85.

Qwen3.5-0.8B 80

Memory earns the full 30, since LM Studio's Q4_K_M build with its projector is 0.73 GB (0.88 GB with the fifth added) and Ollama's Q8_0 build was 1.0 GB, both far within 16 GB. Alibaba publishes no quantised 0.8B, so builds scores 0. Runtimes reach the 25 cap on llama.cpp, Ollama (through its pinned llama.cpp), vLLM, SGLang, LM Studio and Transformers. Apache 2.0 with nothing added gives terms 15 and the ungated repository download 10. Hosted too scores 0, as the Hub maps no inference provider to this size. 30+25+0+15+10+0 = 80.