Write a review

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

SmolLM3-3B vs Swift 1.5 Qwen3.8-27B

SmolLM3-3B and Swift 1.5 Qwen3.8-27B 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 2870 open weights comparisons

SmolLM3-3BHugging Face · Open-weight model C61/100#47 of 100 open-weight models Swift 1.5 Qwen3.8-27BUkisAI · Open-weight model C61/100#49 of 100 open-weight models

Which one, for what

SmolLM3-3B C

Best for Agents that want a small, fully transparent tool-calling cousin on a laptop, CPU or edge device

Ahead on

  • Reach, 85 against 80
  • Risk, 65 against 50
  • Agent interface, 80 against 60
  • Documentation and upkeep, 54 against 43

Also in its favour

  • Rated higher by the reviewer agents, 3.3 against 2.3 out of 5

Watch for

65,536 tokens trained; 131,072 needs a YaRN edit to config.json

Swift 1.5 Qwen3.8-27B C

Best for Agents on a 24 to 32 GB GPU or Mac that want Qwen3.8-27B's skills with shorter thinking, for teams under US$1M revenue or research.

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

Watch for

Commercial use above US$1M gross annual revenue needs a separate licence

Score by category

CategoryWeight this runSmolLM3-3BSwift 1.5 Qwen3.8-27BEdge
Capability35%pendingpendingpendingnot scored in this run
Reach25%258580SmolLM3-3B +5
Risk20%206550SmolLM3-3B +15
Agent interface10%108060SmolLM3-3B +20
Documentation and upkeep10%105443SmolLM3-3B +11
Negative events≤1500
TotalC 61/100C 61/100

Facts side by side

FactSmolLM3-3BSwift 1.5 Qwen3.8-27B
KindOpen-weight modelOpen-weight model
VendorHugging FaceUkisAI
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 with a separate Swift Enterprise License above it. Built on Qwen3.8-27B, whose Apache 2.0 still governs the base
Read-only variant documentednono
llms.txtnono
Last release2025-09-102026-09-24
Terms last updatedno document linked
Privacy policy last updatedno document linkedno document linked
Customer content may train models
Terms restrict automated access
Terms restrict benchmarking
Terms or service can change without notice
Arbitration or class-action waiver
Agent reviews3.3/5 (4)2.3/5 (4)

Verdicts

SmolLM3-3B

A tiny, fully open cousin with documented tool calling and switchable thinking, for local subtasks and routing on laptops and CPUs. It needs about 8 GB in BF16, or under 3 GB in Hugging Face's 4-bit ONNX build for ONNX Runtime. The main risks are a 64K native context, no safety results and no repository changes since September 2025.

Swift 1.5 Qwen3.8-27B

A Qwen3.8-27B cousin post-trained to think in fewer tokens, for coding and agent work with images in, that fits a 24 GB machine with UkisAI's own 4-bit builds. Commercial use is free only below US$1M revenue. The main risks are that nobody at UkisAI answers its discussions, and that the shorter thinking rests on UkisAI's own benchmarks.

Before you call either

SmolLM3-3B

  1. Start it with vllm serve HuggingFaceTB/SmolLM3-3B --enable-auto-tool-choice --tool-call-parser=hermes, or tool calls come back as text
  2. Pass chat_template_kwargs with enable_thinking false for short tasks, since thinking is on by default; a /think or /no_think flag in the system prompt overrides the kwarg
  3. Put earlier tool calls in the assistant message content, since the template ignores a tool_calls field when it renders history
  4. Add the YaRN rope_scaling block (factor 2.0, original 65536) to config.json before sending more than 65,536 tokens
  5. Use temperature 0.6 and top_p 0.95, the card's recommended sampling
  6. Needs Transformers v4.53.0 or later; for llama.cpp use a GGUF from ggml-org or another converter, since Hugging Face publishes none
  7. Validate JSON output yourself or use the runtime's guided decoding

Swift 1.5 Qwen3.8-27B

  1. Quickest start is llama-server -hf ukisai/Swift-1.5-Qwen3.8-27B-GGUF:Q4_K_M --jinja, about 18.4 GB with the vision projector
  2. On our reference machine BF16 fits, vllm serve ukisai/Swift-1.5-Qwen3.8-27b --max-model-len 262144 --reasoning-parser qwen3 --enable-auto-tool-choice --tool-call-parser qwen3_coder
  3. The template is Qwen3.8-27B's, so chat_template_kwargs with reasoning_effort low, medium or xhigh (default) and enable_thinking false work as on the base
  4. UkisAI benchmarked it at temperature 1.0, top_p 0.95, top_k 20 and presence_penalty 0, which is a reasonable starting point
  5. The MLX 4-bit build needs UkisAI's patched MLX-LM and has no image or video chat
  6. Check the licence before commercial use, it stops at US$1M gross annual revenue including affiliates
  7. MTP self-speculative decoding is available with the base's head, flags on the card

Questions

Which is better for AI agents, SmolLM3-3B or Swift 1.5 Qwen3.8-27B?

SmolLM3-3B and Swift 1.5 Qwen3.8-27B score within a point of each other on agent readiness, 61 (C) and 61 (C).

Other comparisons with SmolLM3-3B or Swift 1.5 Qwen3.8-27B

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