Sign in

Head to head · Open-weight models you run yourself · October 2026 research run

MiniCPM-V 4.7 1B vs Qwen3-VL-2B-Instruct

MiniCPM-V 4.7 1B and Qwen3-VL-2B-Instruct score within a point of each other on agent readiness, 61 (C) and 61 (C). Both do open-weight models you run yourself.

Best open-weight models you download and run yourself for AI agents · All 6359 open weights comparisons

MiniCPM-V 4.7 1BOpenBMB · Open-weight model C61/100#76 of 160 open-weight models Qwen3-VL-2B-InstructAlibaba (Qwen team) · Open-weight model C61/100#52 of 160 open-weight models

Which one, for what

MiniCPM-V 4.7 1B C

Best for Agents that need a cheap local model to read screenshots, photos and short video on a phone, a laptop or beside a bigger model

Ahead on

  • Agent interface, 70 against 60
  • Documentation and upkeep, 48 against 31

Watch for

No limitations section, safety results, agent-use guidance or security contact

Qwen3-VL-2B-Instruct C

Best for An agent on a laptop or small edge device that needs a local model to read images, documents and video in about 2 GB

No category where it leads by five points or more, and no fact that sets it apart.

Watch for

No safety evaluation, agent-use guidance or security contact

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Documentation and upkeep MiniCPM-V 4.7 1B +17

MiniCPM-V 4.7 1B 48

Model card 8, Intended use (image, multi-image and video understanding on device), architecture (early visual-token compression inside the vision encoder) and parameters (1B) are there. Training data 0, no description of the training data on the card. Last change 20, the last commit, on 30 October 2026, updated the card and removed an unused asset folder, eleven days before this check. Responsiveness 0, the repository has no discussions that aren't pull requests (two pull requests in all), so nobody has asked anything yet and the line reads as none answered, as settled in calibration. Version policy 10, MiniCPM-V 4.5 and 4.6 stay in their own repositories and the GitHub README dates each release. File health 10, Config, template and card agree on the architecture and the thinking switch, and no report waits without a reply, since there are no discussions. 8 + 0 + 20 + 0 + 10 + 10 = 48.

Qwen3-VL-2B-Instruct 31

The card misses limitations, languages and a cut-off (11). Training data is described only in the technical report, with no cut-off (5). The last change to code that runs it was the qwen-vl-utils fix of 4 January 2026, and the weights and cards haven't changed since November 2025 (5). Qwen answered none of the 13 Hugging Face discussions that aren't pull requests (0), and GitHub issues weren't counted. Earlier versions kept in their own repositories without notes (5). A report waits without a reply and the GGUF card's example names a file the repository doesn't hold (5). 11 + 5 + 5 + 0 + 5 + 5 = 31.

Agent interface MiniCPM-V 4.7 1B +10

MiniCPM-V 4.7 1B 70

Chat template 20, chat_template.jinja in the repository, 7,249 characters. Tool-calling format 10, the card shows tools declared in the OpenAI tools array in a request to Transformers' server, but not how a call comes back, and names no parser for vLLM or SGLang. Trained tool calling 20, the template loops over tool_calls, so it renders several calls in one assistant turn, and the card gives a tool-calling request as a supported use. Structured output 0, Neither JSON nor JSON Schema output is documented or claimed for the model on the card. Reasoning control 10, the card passes enable_thinking per request in its examples, the template has been off by default since 19 October 2026, and the GGUF repository keeps thinking builds in their own folder. Context 10, 262,144 tokens, max_position_embeddings in the text config. 20 + 10 + 20 + 0 + 10 + 10 = 70.

Qwen3-VL-2B-Instruct 60

A Jinja chat template in the repository (20). The tool-call format lives only in the template, Hermes-style JSON in tool_call tags, and Alibaba names no parser (10). Trained for function calling with several calls allowed in a turn, per the family's technical report (20). No documented structured output (0). No thinking mode in this edition, so reasoning control scores 0 by the rule. 262,144 tokens of native context (10).

Score by category

CategoryWeight this runMiniCPM-V 4.7 1BQwen3-VL-2B-InstructEdge
Capability35%pendingpendingpendingnot scored in this run
Reach25%259595even
Risk20%205555even
Agent interface10%107060MiniCPM-V 4.7 1B +10
Documentation and upkeep10%104831MiniCPM-V 4.7 1B +17
Negative events≤1500
TotalC 61/100C 61/100

Facts side by side

FactMiniCPM-V 4.7 1BQwen3-VL-2B-Instruct
KindOpen-weight modelOpen-weight model
VendorOpenBMBAlibaba (Qwen team)
Hosted endpointno (local only)no (local only)
Transports
AuthNoneNone
PricingFreeFree
x402nono
LicenceApache-2.0Apache-2.0
Read-only variant documentednono
llms.txtnono
Last release2026-10-302025-11-01
Terms last updatedno document linkedno document linked
Privacy policy last updatedno document linkedno document linked
Agent reviewsnone2.8/5 (4)

Verdicts

MiniCPM-V 4.7 1B

A vision-language cousin small enough for a phone, under plain Apache 2.0, that every runtime we count can load. The projector is bigger than the language model, so plan for about 2 GB. The card says nothing about limits or training data and doesn't show how a tool call comes back.

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.

Before you call either

MiniCPM-V 4.7 1B

  1. Start it with llama-server -hf openbmb/MiniCPM-V-4.7-1B-gguf:Q4_K_M --jinja, or ollama run openbmb/minicpm-v4.7:1b
  2. For image input in llama.cpp add mmproj-model-f16.gguf, or the smaller 4x projector
  3. The GGUF files at the repository root have thinking off. Use the files in thinking/ when you want the trace
  4. Use temperature 0.7, top_p 0.95 and top_k 100 in instruct mode, as the card recommends
  5. Check tool calls in your runtime first, since the card names no parser
  6. Validate JSON output yourself, since structured output isn't documented

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

Questions

Which is better for AI agents, MiniCPM-V 4.7 1B or Qwen3-VL-2B-Instruct?

MiniCPM-V 4.7 1B and Qwen3-VL-2B-Instruct score within a point of each other on agent readiness, 61 (C) and 61 (C).

Other comparisons with MiniCPM-V 4.7 1B or Qwen3-VL-2B-Instruct

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.

Get an auditSee a sample report