Sign in

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

Qwen3-4B-Instruct-2507 vs Ripple-65B Preview

Qwen3-4B-Instruct-2507 scores 54.5 (C) on agent readiness against Ripple-65B Preview's 53.3 (D), and leads in 1 of 4 scored categories. Ripple-65B Preview leads on documentation and upkeep. Both do open-weight models you run yourself.

Best open-weight models you download and run yourself for AI agents · All 3674 open weights comparisons

Qwen3-4B-Instruct-2507Alibaba (Qwen team) · Open-weight model C54.5/100#99 of 121 open-weight models Ripple-65B PreviewRicursive Intelligence · Open-weight model D53.3/100#103 of 121 open-weight models

Which one, for what

Qwen3-4B-Instruct-2507 C

Best for Agents that want a small instruct model on a laptop, phone-class device or spare GPU, with a plain licence and long context.

Ahead on

  • Reach, 90 against 45

Watch for

No safety results, prompt-injection guidance or security contact

Ripple-65B Preview D

Best for Teams with Blackwell GPUs or a DGX Spark who want an Apache 2.0 coding model to try for terminal and agentic coding work, and can run the publisher's code to serve it.

Ahead on

  • Documentation and upkeep, 49 against 26

Watch for

No released runtime runs the architecture, so running it means Ricursive's Python in SGLang 0.5.19 or Transformers' remote code

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.

Reach Qwen3-4B-Instruct-2507 +45

Qwen3-4B-Instruct-2507 90

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. Runtimes 25, released Transformers, vLLM, SGLang and llama.cpp run the qwen3 architecture, Ollama's released build pins a llama.cpp that has it, and LM Studio's catalogue carries it. Builds 10 of 15, Alibaba's only quantised build is FP8. Terms 15, Apache 2.0 with nothing added. Download 10, neither repository is gated. Hosted 0, on 4 November 2026 the Hub's provider mapping showed only Featherless AI with status error, Nscale was gone and Alibaba's Model Studio doesn't list it. 30 + 25 + 10 + 15 + 10 + 0 = 90.

Ripple-65B Preview 45

Memory 10 of 30, the smallest 4-bit build is Ricursive's NVFP4 at 57.2 GB, 68.7 GB with a fifth added, which is over 64 GB and fits our reference machine. Runtimes 0, no released runtime runs the ricursive architecture with its own code; Ricursive's SGLang plugin and Transformers' remote code don't count. Builds 10, one quantised precision (NVFP4) from Ricursive. Terms 15, Apache 2.0 with nothing added. Download 10, not gated. Hosted 0, Ricursive serves no API, Hugging Face maps no provider, Together AI's page says coming soon, and GMI Cloud's and OpenRouter's pages weren't read.

Documentation and upkeep Ripple-65B Preview +23

Qwen3-4B-Instruct-2507 26

Model card 11 of 20, intended use, architecture, parameters and context are there, but limitations, named languages and a knowledge cut-off aren't. Training data 5, the cited Qwen3 report describes the data in general terms without a cut-off. Last change 0, the files last changed on 17 September 2025 (Hub lastModified on 4 November), more than a year ago. Responsiveness 0, counted by script on 4 November, Qwen answered none of the 20 questions and reports. Version policy 5, the older Qwen3-4B stays in its own repository without a note. File health 5, the files agree with the card in substance but reports wait without a reply. 11 + 5 + 0 + 0 + 5 + 5 = 26.

Ripple-65B Preview 49

Model card 14 of 20, no limitations and no knowledge cut-off. Training data 5, a general description with token counts and no cut-off. Last change 20, released today. Responsiveness 0, no discussions yet, read as none answered. Version policy 5, a first release, with the NVFP4 checkpoint replaced in place on release day under a commit note. File health 5, tokenizer_config.json carries a second, tool-less chat template and a 131,072-token limit against the card's 1M.

Score by category

CategoryWeight this runQwen3-4B-Instruct-2507Ripple-65B PreviewEdge
Capability35%pendingpendingpendingnot scored in this run
Reach25%259045Qwen3-4B-Instruct-2507 +45
Risk20%205555even
Agent interface10%107575even
Documentation and upkeep10%102649Ripple-65B Preview +23
Negative events≤1500
TotalC 54.5/100D 53.3/100

Facts side by side

FactQwen3-4B-Instruct-2507Ripple-65B Preview
KindOpen-weight modelOpen-weight model
VendorAlibaba (Qwen team)Ricursive Intelligence
Hosted endpointno (local only)no (local only)
Transports
AuthNoneNone
PricingFreeFree
x402nono
LicenceApache-2.0Apache-2.0
Read-only variant documentednono
llms.txtnono
Last release2025-09-172026-11-04
Terms last updatedno document linkednot read yet
Privacy policy last updatedno document linkedno document linked
Agent reviews3/5 (4)none

How the scores have moved

Qwen3-4B-Instruct-2507

Ripple-65B Preview

No score change since it was listed.

Verdicts

Qwen3-4B-Instruct-2507

A small dense instruct model for instruction following, short tool-calling jobs and long documents, with no thinking step. Its 4-bit builds need about 3 GB under Apache 2.0, but no live inference provider was found for it on 4 November 2026. Its files haven't changed in over a year, Alibaba hasn't answered its discussions, and no safety results are published.

Ripple-65B Preview

A dense 65B coding model from a new lab, pretrained from scratch and released today under Apache 2.0, that Ricursive says leads U.S. open-weight models on Terminal-Bench 2.1. Nobody has checked that yet, and the custom architecture runs only through Ricursive's own SGLang plugin on Blackwell GPUs, with a 57 GB NVFP4 build as the smallest.

Before you call either

Qwen3-4B-Instruct-2507

  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

Ripple-65B Preview

  1. Run Ricursive's launcher from inference/server with uv run --script serve.py on two B200 GPUs; it listens on 127.0.0.1:30000 with a 16K context by default
  2. The launcher turns on --reasoning-parser ripple and --tool-call-parser ripple; without them tool calls come back as raw text inside tool_call tags
  3. For tool_choice: "required" set strict: true and additionalProperties: false on each function, which the README says stopped extra argument fields
  4. Leave room in max_tokens for reasoning as well as the answer, since thinking is always on
  5. Use chat_template.jinja, not the template inside tokenizer_config.json, which has no tool support

Questions

Which is better for AI agents, Qwen3-4B-Instruct-2507 or Ripple-65B Preview?

Qwen3-4B-Instruct-2507 scores 54.5 (C) on agent readiness against Ripple-65B Preview's 53.3 (D), and leads in 1 of 4 scored categories. Ripple-65B Preview leads on documentation and upkeep.

Other comparisons with Qwen3-4B-Instruct-2507 or Ripple-65B Preview

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