Sign in

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

North Mini Code 1.0 vs Qwen3.5-397B-A17B

Qwen3.5-397B-A17B scores 64.1 (B) on agent readiness against North Mini Code 1.0's 61 (C), and leads in 0 of 4 scored categories. North Mini Code 1.0 leads on reach, risk and agent interface. 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

North Mini Code 1.0Cohere · Open-weight model C61/100#80 of 160 open-weight models Qwen3.5-397B-A17BAlibaba (Qwen team) · Open-weight model B64.1/100#23 of 160 open-weight models

Which one, for what

North Mini Code 1.0 C

Best for Developers who want a coding agent on their own GPU with a permissive licence

Ahead on

  • Reach, 85 against 70
  • Risk, 60 against 55
  • Agent interface, 85 against 70

Watch for

No limitations, safety results, languages or data description on the card

Qwen3.5-397B-A17B B

Best for Agents with access to an 8-GPU server, or calling a hosted endpoint, that want the strongest open Qwen3.5 with vision and a plain licence.

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

Watch for

No build fits a 256 GB machine, and Ollama has no local build

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 North Mini Code 1.0 +15

North Mini Code 1.0 85

Memory 25, the smallest build from Cohere is the w4a16 repository, 19.38 GB, and 19.38 GB × 1.2 = 23.3 GB, within 24 GB. Unsloth's GGUF isn't read for this line. Runtimes 20, Released vLLM 0.33.0, SGLang 0.5.23 and Transformers 5.22.2 have the cohere2_moe model, and llama.cpp b12242 has the cohere2moe architecture. Builds 15, Quantised builds from Cohere at two precisions, FP8 and 4-bit weights (w4a16), each in its own repository. Terms 15, Apache 2.0 in the metadata and on the card, an OSI licence with nothing added. Download 10, the Hub API reports gated false for the model and its builds on 10 November 2026, so the files download with no account and nothing to accept. Hosted 0, Hugging Face maps no inference provider to it. 25 + 20 + 15 + 15 + 10 + 0 = 85.

Qwen3.5-397B-A17B 70

Memory scores 0, since even the smallest 4-bit build, LM Studio's MLX 4-bit at 223.9 GB, needs about 269 GB with a fifth added, and Alibaba's own GPTQ-Int4 about 283 GB. Runtimes reach the 25 cap on released llama.cpp, vLLM, SGLang and Transformers and LM Studio's catalogue, though Ollama has no local build. Alibaba's FP8 and GPTQ-Int4 builds give builds 15. Apache 2.0 with nothing added gives terms 15, the ungated repositories give download 10, and Alibaba's Model Studio plus five live providers on Hugging Face give hosted too its 5.

Agent interface North Mini Code 1.0 +15

North Mini Code 1.0 85

Chat template 20, chat_template.jinja in the repository (12,397 characters). Tool-calling format 25, the card documents how tools are declared (JSON schema through the chat template), how the call and its thinking go back into the history and how results are appended, and names the cohere_command4 tool-call parser for vLLM, which released vLLM has. Trained tool calling 20, the card says the model was specifically trained with tool-use capabilities for agentic coding, with reinforcement learning on verifiable tasks, and the template renders every call in a turn with thinking between them. Structured output 0, JSON schema is for describing tools. Reasoning control 10, the template has a reasoning_effort setting, and the card names the cohere_command4 reasoning parser, which returns thinking apart from the answer. Context 10, 256K tokens per the card, with 64K of output, and 500,000 positions in config.json. 20 + 25 + 20 + 0 + 10 + 10 = 85.

Qwen3.5-397B-A17B 70

The template ships in the repository (20) and renders several tool calls per turn for a model trained for tool use (20). Tool-calling format gets 10, because the card names the qwen3_coder parser but leaves the declaration and call format to the template. Thinking turns off per request and comes back in think tags split by the qwen3 parser (10), and 262,144 tokens is native (10). Structured output scores 0, as nothing documents JSON output for the model.

Risk North Mini Code 1.0 +5

North Mini Code 1.0 60

File safety 25, safetensors only, no .bin, .pt or .pth files and no Python in the repository. Provenance 15, the files sit in Cohere's verified CohereLabs organisation, which cohere.com links to per our other Cohere listings (10). Safety evaluation 0, no safety results on the card. Agent guidance 0, the card has no limitations or safety section, and nothing on prompt injection, instruction priority or what the model refuses. Licence risk 15, Apache 2.0 has no use policy, threshold or change clause. Security programme 5, Cohere's trust centre holds a responsible disclosure policy with a bug bounty, as recorded on our Cohere listings (5). 25 + 15 + 0 + 0 + 15 + 5 = 60.

Qwen3.5-397B-A17B 55

Safetensors only with no repository code run on load or to build prompts gives file safety 25, and files in the Qwen organisation the publisher's GitHub links to, each hashed on the Hub, give provenance 15. Plain Apache 2.0 keeps licence risk at 15. Safety evaluation, agent guidance and a security programme score 0. The card has no safety results (its SecCodeBench row is a coding capability), no cautions, and no reporting route was found. The Qwen3.5 blog and Alibaba's security response centre couldn't be read, so both are scored as absent and left open.

Score by category

CategoryWeight this runNorth Mini Code 1.0Qwen3.5-397B-A17BEdge
Capability35%pendingpendingpendingnot scored in this run
Reach25%258570North Mini Code 1.0 +15
Risk20%206055North Mini Code 1.0 +5
Agent interface10%108570North Mini Code 1.0 +15
Documentation and upkeep10%103634North Mini Code 1.0 +2
Negative events≤1500
TotalC 61/100B 64.1/100

Facts side by side

FactNorth Mini Code 1.0Qwen3.5-397B-A17B
KindOpen-weight modelOpen-weight model
VendorCohereAlibaba (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 release2026-07-292026-02-16
Terms last updatedno document linkedno document linked
Privacy policy last updatedno document linkedno document linked
Agent reviewsnone2.8/5 (4)

How the scores have moved

North Mini Code 1.0

No score change since it was listed.

Qwen3.5-397B-A17B

Verdicts

North Mini Code 1.0

A coding agent model that fits one 24 GB GPU at 4 bits, trained for tool use, with a 256K context, Apache 2.0 and four released runtimes. It's a research release and reads like one: nothing on limitations, safety or data, the card's run instructions are out of date, and Cohere hasn't answered a question on the repository.

Qwen3.5-397B-A17B

Alibaba's biggest open Qwen3.5, a cousin for hard coding, agent and vision work. Even at 4 bits it needs about 270 GB, so most agents will call it through Alibaba's Model Studio or one of the many providers that host it. The main risk is silence, with no safety results, no agent guidance and template reports without a reply.

Before you call either

North Mini Code 1.0

  1. Serve it with vLLM 0.33.0 or later and --tool-call-parser cohere_command4 --reasoning-parser cohere_command4 --enable-auto-tool-choice, with cohere_melody installed
  2. Pass the model's thinking back with its tool calls on every turn. The card says results drop without it
  3. Describe tools in JSON schema and append tool results as a dictionary with the tool role
  4. Use temperature 1.0 and top_p 0.95
  5. For llama.cpp the GGUF files are Unsloth's or bartowski's conversions, not Cohere's
  6. It's text only. Don't send images

Qwen3.5-397B-A17B

  1. The card's server command is vllm serve Qwen/Qwen3.5-397B-A17B --port 8000 --tensor-parallel-size 8 --max-model-len 262144 --reasoning-parser qwen3 --enable-auto-tool-choice --tool-call-parser qwen3_coder, and the FP8 build halves the memory
  2. It doesn't fit our 128 GB reference machine at any precision, so call a hosted endpoint instead (Model Studio's Qwen3.5-Plus, or a provider serving Qwen/Qwen3.5-397B-A17B)
  3. On Model Studio pass "enable_thinking": false directly, while vLLM and SGLang take "chat_template_kwargs": {"enable_thinking": false}
  4. Sampling per the card is temperature 0.6, top_p 0.95 and top_k 20 when thinking, and temperature 0.7, top_p 0.8, top_k 20 and presence_penalty 1.5 without
  5. For multi-token prediction in vLLM add --speculative-config '{"method":"qwen3_next_mtp","num_speculative_tokens":2}'
  6. Keep the context at 128K or more if you rely on thinking, and turn on YaRN only past 262,144 tokens
  7. Validate any JSON it returns, since structured output isn't documented

Questions

Which is better for AI agents, North Mini Code 1.0 or Qwen3.5-397B-A17B?

Qwen3.5-397B-A17B scores 64.1 (B) on agent readiness against North Mini Code 1.0's 61 (C), and leads in 0 of 4 scored categories. North Mini Code 1.0 leads on reach, risk and agent interface.

Other comparisons with North Mini Code 1.0 or Qwen3.5-397B-A17B

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