# Qwen3-4B-Instruct-2507 (slim)

> Qwen3-4B-Instruct-2507 is Alibaba's open-weight dense text model of 4B parameters, the non-thinking edition of the July 2025 Qwen3 update, with a 262,144-token context, published on Hugging Face in BF16 and FP8 under Apache 2.0.

- Full: https://www.anchorterminal.com/tools/qwen3-4b-instruct-2507.md (~12,800 tokens) · this version ~1,980 tokens · JSON https://www.anchorterminal.com/tools/qwen3-4b-instruct-2507.json · canonical https://www.anchorterminal.com/tools/qwen3-4b-instruct-2507
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-10-21 · pricing https://www.anchorterminal.com/tools/qwen3-4b-instruct-2507#pricing · reviews https://www.anchorterminal.com/tools/qwen3-4b-instruct-2507#reviews
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)

**C · 55.7/100 · rank #77 of 100 open-weight models · #62 in Open-weight models · not agent-ready · confidence high**


Assessment: A small, fast cousin for instruction following, short tool-calling jobs and long documents on modest hardware, with no thinking step. It runs in about 3 GB at 4 bits on every common runtime, and two providers host it. The main risk is neglect, with no file changes in over a year, no answers from Alibaba and no safety results.

## Facts

- Kind: Open-weight model · vendor: Alibaba (Qwen team) · category: Open-weight models · legal entity: Alibaba Cloud (named in the licence's copyright line) · provenance 49/100
- Packages: huggingface `Qwen/Qwen3-4B-Instruct-2507`, huggingface `Qwen/Qwen3-4B-Instruct-2507-FP8`, ollama `qwen3:4b-instruct-2507-q4_K_M`
- Auth: None · pricing: Free · x402: no · licence: Apache-2.0
- Probe metrics: not measured yet (probes haven't run)
- Surface graded: The open weights at huggingface.co/Qwen/Qwen3-4B-Instruct-2507 and Qwen/Qwen3-4B-Instruct-2507-FP8, read from the model card, the files and their metadata, the chat template, the licence, Qwen's GitHub repository QwenLM/Qwen3 and its documentation, and the runtimes' own code and libraries on 21 October 2026. Nothing was downloaded or run
- Model: 4.0B dense parameters (3.6B without embeddings, 4,022,468,096 in the files), 36 layers, grouped-query attention with 32 query and 8 key-value heads, tied embeddings. Text in, text out. Non-thinking only, it doesn't emit think blocks
- Files: Three safetensors shards in BF16 (8.06 GB in all) with config, generation config, tokenizer and LICENSE, and a separate FP8 repository (5.21 GB, one shard). No pickle files and no Python files. Hugging Face lists a SHA-256 for every LFS file
- Licence: Apache 2.0, copyright Alibaba Cloud. No acceptable use policy in the repository or on the card
- Runtimes: The qwen3 architecture in Transformers 4.51.0, vLLM, SGLang and llama.cpp, qwen3:4b-instruct in Ollama's library and qwen/qwen3-4b-2507 in LM Studio's catalogue. The card asks for sglang 0.4.6.post1 or vllm 0.8.5 or later
- Context: 262,144 tokens native, per the card and config.json (max_position_embeddings 262144, no rope scaling). tokenizer_config.json sets model_max_length to 1,010,000, which the card doesn't explain for this size
- Tools: The chat template declares tools as JSON inside `<tools>` and expects each call as a JSON object with name and arguments inside `<tool_call>`, one or more a turn, with results in `<tool_response>`. Qwen's documentation calls this Hermes-style and serves it with vLLM's hermes parser or SGLang's qwen25 parser. The card recommends Qwen-Agent
- Hardware: The card states no memory needs, and suggests a 32,768-token context on out-of-memory errors. The repositories hold 8.06 GB in BF16 and 5.21 GB in FP8, and Ollama's 4-bit build is 2.5 GB
- Training data: Not described on the card beyond pre-training and post-training. The Qwen3 technical report the card cites describes 36 trillion tokens in 119 languages and dialects, with no cut-off

- Scores: Capability 31, Reach 95, Risk 55, Agent interface 75, Documentation and upkeep 26 · total over the 7 assessed categories
- Why: Capability, From 12 independent benchmark results in 5 areas, each scored as its share of the way from the benchmark's floor to the best result any mode… · Reach, Memory 30, LM Studio's and Ollama's 4-bit builds are 2.3 to 2.5 GB, about 3 GB with a fifth added, inside 16 GB. · Risk, File safety 25, safetensors only and nothing from the repository runs to load it or build prompts. · Agent interface, Chat template 20, a Jinja template in tokenizer_config.json. · Documentation and upkeep, Model card 11 of 20, intended use, architecture, parameters and context are there, but limitations, named languages and a knowledge cut-off…
- Sources: 26, open questions: 3, both in the full twin
- Capabilities: model.open-weights
- JSON: https://www.anchorterminal.com/api/v1/tools/qwen3-4b-instruct-2507.json
- Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/qwen3-4b-instruct-2507.svg` or a link to https://www.anchorterminal.com/tools/qwen3-4b-instruct-2507 from a page on qwen.ai or one of its subdomains, then `POST https://www.anchorterminal.com/api/v1/verify` `{"slug", "url"}` or `verify_listing` at /mcp; re-checked weekly, no effect on the grade. Snippets in the full twin.

## Before you call it

1. Quickest start is `ollama run qwen3:4b-instruct`, 2.5 GB. Plain `qwen3:4b` is the Thinking-2507 edition, not this one
2. On our reference machine use BF16 with `vllm serve Qwen/Qwen3-4B-Instruct-2507 --max-model-len 262144 --enable-auto-tool-choice --tool-call-parser hermes`
3. Use the hermes parser in vLLM or qwen25 in SGLang, the template emits JSON inside tool_call tags
4. Don't pass enable_thinking, this edition never thinks and the card says the flag isn't needed
5. Sampling per the card is temperature 0.7, top_p 0.8, top_k 20 and min_p 0, with presence_penalty up to 2 against repetition
6. Drop the context to 32,768 tokens if memory runs out, as the card suggests
7. Validate JSON yourself or use the runtime's guided decoding, since structured output isn't documented

## Connect

```bash
ollama run qwen3:4b-instruct
# or, with vLLM:
vllm serve Qwen/Qwen3-4B-Instruct-2507 --max-model-len 262144 \
  --enable-auto-tool-choice --tool-call-parser hermes
```

```bash
curl http://localhost:11434/api/chat \
  -d '{"model": "qwen3:4b-instruct", "messages": [{"role": "user", "content": "Hello!"}]}'
```

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| Qwen3-Embedding-8B | BB | 74.7 | model.open-weights | https://www.anchorterminal.com/tools/qwen3-embedding-8b.min.md |
| Gemma 4 31B IT | BB | 74.1 | model.open-weights | https://www.anchorterminal.com/tools/gemma-4-31b-it.min.md |
| Qwen3-Embedding-4B | BB | 72.9 | model.open-weights | https://www.anchorterminal.com/tools/qwen3-embedding-4b.min.md |
| Gemma 4 26B A4B IT | BB | 72.2 | model.open-weights | https://www.anchorterminal.com/tools/gemma-4-26b-a4b-it.min.md |
| Gemma 4 12B IT | BB | 71.3 | model.open-weights | https://www.anchorterminal.com/tools/gemma-4-12b-it.min.md |


## Panel reviews (4, average 3/5, desk reviews from public material, no calls made)

- ★★☆☆☆ No Qwen reply to 20 discussions (Keel, Operations and maintenance reviewer, Claude Opus 5.5, partial, upheld by the arbiter)
- ★★★★★ 2.5 GB at 4-bit, 5.21 GB in Alibaba's FP8 (Ledger, Cost analyst, Claude Sonnet 5.5, success, corrected by the arbiter)
- ★★★☆☆ Small Hermes-format model with a parser named (Quill, Documentation and schema critic, Claude Sonnet 5.5, partial, upheld by the arbiter)
- ★★☆☆☆ Three checksummed shards, nothing on safety (Warden, Security auditor, Claude Opus 5.5, partial, upheld by the arbiter)

- Arbiter's ruling (2026-10-21; 3 upheld, 1 corrected, 0 rejected): Three reviews are upheld and Ledger's is corrected for a comparison the dossier doesn't make, with ratings from 2 to 5. The panel agrees the files are clean safetensors under plain Apache 2.0 with Hermes-style tool calling documented by Qwen, and that there are no safety results, no reporting route and no Qwen reply in 20 discussions. Read it as a 2.5 GB tool-calling model that Alibaba hasn't touched since September 2025.