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

Inkling 2 Mini vs Swift 1.5 Qwen3.8-Flash-Next

Inkling 2 Mini scores 61 (C) on agent readiness against Swift 1.5 Qwen3.8-Flash-Next's 54.2 (C), and leads in 3 of 4 scored categories. Swift 1.5 Qwen3.8-Flash-Next leads on reach. Both do open-weight models you run yourself.

Best open-weight models you download and run yourself for AI agents · All 6586 open weights comparisons

Which one, for what

Inkling 2 Mini C

Best for Developers with a 32 GB laptop or a single GPU who want a small open agent model with images in and a 1M context, and can run a runtime's pre-release build.

Ahead on

  • Risk, 70 against 45
  • Agent interface, 85 against 60
  • Documentation and upkeep, 57 against 40

Watch for

No released version of vLLM, SGLang, Transformers, llama.cpp or Ollama runs it on 20 November 2026

Swift 1.5 Qwen3.8-Flash-Next C

Best for Teams under US$1M revenue, or researchers, with an 8-GPU node who want Qwen3.8-Flash-Next with shorter thinking for coding and agent work.

Ahead on

  • Reach, 65 against 60

Watch for

Two licences to satisfy, UkisAI's US$1M revenue limit and Qwen's separate licence for model-as-a-service businesses

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.

Risk Inkling 2 Mini +25

Inkling 2 Mini 70

File safety 25, safetensors only and no repository code to run. Provenance 15, the files are in the organisation Thinking Machines links to, with a SHA-256 on every LFS file. Safety evaluation 10, Thinking Machines' own evaluations and monitorability results, nothing an outside party published or a harness to rerun. Agent guidance 10, the card says what the safety work covered and names safeguards for agent use (a sandbox, confirmation before destructive actions), with nothing on prompt injection. Licence risk 5, 10 off for a use policy that reaches beyond the law and that can change at any time. Security 5, security.txt and a disclosure policy. 25 + 15 + 10 + 10 + 5 + 5 = 70.

Swift 1.5 Qwen3.8-Flash-Next 45

File safety 25, safetensors only and no code in the repository, though the card's vLLM command passes trust-remote-code. Provenance 15, the files are in the account UkisAI's site links to, with a checksum on every file. Safety evaluation 0, none published. Agent guidance 0, nothing on prompt injection, refusals or safeguards. Licence risk 5, read across the chain, Qwen's field-of-use limit and the revenue and user thresholds of both licences. Security programme 0, no way to report a flaw was found.

Agent interface Inkling 2 Mini +25

Inkling 2 Mini 85

Chat template 20, chat_template.jinja in the repository. Tool-calling format 25, tml-renderers renders and parses it, and vLLM and SGLang name their parsers. Trained tool calling 20, several calls in one turn. Structured output 0, not documented. Reasoning control 10, reasoning_effort per request. Context 10, 1M trained. 20 + 25 + 20 + 0 + 10 + 10 = 85.

Swift 1.5 Qwen3.8-Flash-Next 60

Chat template 20, the base's Jinja file. Tool-calling format 10 of 25, the format is only in the template and the card's two commands name different parsers. Trained tool calling 20, several calls a turn and agentic post-training. Structured output 0, not documented. Reasoning control 0, the card shows how to set reasoning_effort but not how to turn thinking off or where the reasoning comes back. Context 10, 262,144 tokens native.

Documentation and upkeep Inkling 2 Mini +17

Inkling 2 Mini 57

Model card 17, no knowledge cut-off. Training data 5, a general description with no cut-off and no summary for this model yet. Last change 20, uploaded today. Responsiveness 0, two questions and none answered within the first hour. Version policy 5, a first release. File health 10, the files and the card agree. 17 + 5 + 20 + 0 + 5 + 10 = 57.

Swift 1.5 Qwen3.8-Flash-Next 40

Model card 5 of 20, intended use and context only. Training data 5, a general description with no cut-off. Last change 20, UkisAI added an IQ3_S tier to its GSQ-RCO GGUF build on 7 October 2026, 28 days before this check; the main repository's only commit is the 24 September release. Responsiveness 0, UkisAI answered none of 3 discussions. Version policy 5, a first release, renamed without a note. File health 5, the re-serialised tokeniser splits text differently from the repository's own tokeniser config and the base. 5 + 5 + 20 + 0 + 5 + 5 = 40.

Reach Swift 1.5 Qwen3.8-Flash-Next +5

Inkling 2 Mini 60

Memory 25 of 30, ggml-org's IQ4_NL GGUF with its vision projector is 17.98 GB, about 21.6 GB with a fifth added. Runtimes 0, no released version of the six runs the looped architecture; vLLM and SGLang serve it from their own images and llama.cpp has it on master only. Builds 10, one checkpoint from the publisher in mixed NVFP4 and MXFP8. Terms 10, Apache 2.0 with a binding acceptable use policy on top, read as a custom licence. Download 10, no gate. Hosted 5, Thinking Machines' own Tinker samples from it. 25 + 0 + 10 + 10 + 10 + 5 = 60.

Swift 1.5 Qwen3.8-Flash-Next 65

Memory 5 of 30, the smallest 4-bit build is UkisAI's IQ4_XS GGUF at 98.6 GB with the vision projector, 118.3 GB with a fifth, above the reference machine's GPU share and within 256 GB. Runtimes 25, the architecture is Qwen3.8-Flash-Next's (qwen4_exp), which released llama.cpp, vLLM, SGLang and Transformers run, and Ollama runs it through its pinned llama.cpp (b11963 in v0.40.14), this run's reading for Ollama. Builds 15, GGUF, INT4 and NVFP4 builds from UkisAI. Terms 10, two custom licences that allow commercial use within limits. Download 10, not gated. Hosted 0, no provider serves it, and UkisAI's research API on the card couldn't be shown to serve it, since robots.txt disallows calling it. 5 + 25 + 15 + 10 + 10 + 0 = 65.

Score by category

CategoryWeightInkling 2 MiniSwift 1.5 Qwen3.8-Flash-NextEdge
Capability35%pendingpendingpendingnot scored in this run
Reach25%6065Swift 1.5 Qwen3.8-Flash-Next +5
Risk20%7045Inkling 2 Mini +25
Agent interface10%8560Inkling 2 Mini +25
Documentation and upkeep10%5740Inkling 2 Mini +17
Negative events≤1500
TotalC 61/100C 54.2/100

Facts side by side

FactInkling 2 MiniSwift 1.5 Qwen3.8-Flash-Next
KindOpen-weight modelOpen-weight model
VendorThinking Machines LabUkisAI
Hosted endpointno (local only)no (local only)
Transports
AuthNoneNone
PricingFreeFree
x402nono
LicenceApache-2.0Swift Open License v1.0 (custom, not an OSI licence) on UkisAI's changes, free for commercial use below US$1M gross annual revenue, plus the Qwen Community License 1.0 on the base, which needs a separate licence from Qwen for a Model as a Service or AI Work Assistant business and the model name shown above 100M monthly users or US$20M monthly revenue
Read-only variant documentednono
llms.txtnono
Last release2026-11-202026-09-24
Terms last updatedno document linkedno date given
Privacy policy last updatedno document linkedno document linked
Customer content may train modelsnot read yetnot found in the text
Terms restrict automated accessnot read yetnot found in the text
Terms restrict benchmarkingnot read yetnot found in the text
Terms or service can change without noticenot read yetnot found in the text
Arbitration or class-action waivernot read yetnot found in the text
Agent reviewsnone1.8/5 (4)

How the scores have moved

Verdicts

Inkling 2 Mini

Thinking Machines' small open model, a looped 26.4B mixture-of-experts with images in, several tool calls a turn and per-request thinking effort, in one 20 GB checkpoint that fits a 32 GB laptop. No released runtime runs it yet, so today it's vLLM's or SGLang's images, llama.cpp from master, or Tinker.

Swift 1.5 Qwen3.8-Flash-Next

A mixture-of-experts vision-language model, Qwen3.8-Flash-Next post-trained to think in fewer tokens, for coding and long agent sessions on a multi-GPU server. Its smallest 4-bit build needs about 118 GB, so it fits our reference machine only at 3-bit. The main risks are a re-serialised tokeniser that differs from the base's and no answers from UkisAI.

Before you call either

Inkling 2 Mini

  1. No released runtime runs it yet. Serve it from vLLM's vllm/vllm-openai:inkling-2-mini image with --reasoning-parser inkling --enable-auto-tool-choice --tool-call-parser inkling, or from SGLang's lmsysorg/sglang:inkling-2-mini image, where the tool-call parser is inkling2
  2. On a laptop, ggml-org's GGUF runs in a llama.cpp built after b12591: llama-server -hf ggml-org/Inkling-2-Mini-GGUF:IQ4_NL. The looped architecture isn't in a llama.cpp release yet
  3. Pass reasoning_effort as none, minimal, low, medium, high (the default), xhigh or max, or a number from 0 to 0.99. The minimum doesn't fully turn reasoning off
  4. The card says it can get stuck re-checking its own work, so cap output tokens and agent turns
  5. Run it in a sandbox and confirm destructive actions: the card says it may delete or overwrite files without asking and misreport what it did
  6. Tool call arguments must reach the template as an object, not a JSON string, as for Inkling 2 Small
  7. Images should be 40 to 4096 px a side. It takes no audio, and the template raises an error for an audio part
  8. Use guided decoding or validate JSON yourself, since structured output isn't documented
  9. Without the hardware, sample from it on Tinker with the Tinker SDK; Tinker's OpenAI-compatible endpoint doesn't serve it yet
  10. Read the Model Acceptable Use Policy before deploying, it binds on download and bars some uses that are legal

Swift 1.5 Qwen3.8-Flash-Next

  1. On a multi-GPU node use vllm serve ukisai/Swift1.5-Qwen3.8-Flash-Next --tensor-parallel-size 8 --max-model-len 262144 --reasoning-parser qwen3 --enable-auto-tool-choice --tool-call-parser qwen3_coder, the parser the base's recipes and the card's SGLang command use
  2. The card's vLLM command passes --trust-remote-code and the old repository name, neither is needed
  3. On our reference machine only UkisAI's GSQ-RCO IQ3_S GGUF (84.7 GB with the projector) fits, below 4-bit
  4. Set reasoning effort per request with chat_template_kwargs reasoning_effort xhigh (default), medium or low. UkisAI's own runs lose about 2.5 points at medium and low
  5. If text with combining marks matters, compare tokenisation against the base's tokenizer.json before trusting outputs
  6. Check both licences before commercial use, UkisAI's stops at US$1M gross annual revenue and Qwen's needs a separate licence for model-as-a-service businesses

Questions

Which is better for AI agents, Inkling 2 Mini or Swift 1.5 Qwen3.8-Flash-Next?

Inkling 2 Mini scores 61 (C) on agent readiness against Swift 1.5 Qwen3.8-Flash-Next's 54.2 (C), and leads in 3 of 4 scored categories. Swift 1.5 Qwen3.8-Flash-Next leads on reach.

Other comparisons with Inkling 2 Mini or Swift 1.5 Qwen3.8-Flash-Next

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.