LLM Reference

North Mini Code 1.0

Released
2026-06-09
Last refreshed
2026-06-09
Status
Researched today
Open SourceCommercial use allowedCodingRAGAgentsLong contextClassificationJSON / Tool useCodingOpen SourceAgents

North Mini Code 1.0 is worth evaluating for coding, rag, and agents when its provider route and context window match the workload.

Use it for

  • Teams evaluating coding, rag, and agents
  • Workloads that can use a 256k context window
  • Buyers comparing 1 tracked provider route

Do not use it for

  • Vision or document-understanding workloads
Specifications
Family
North
Released
2026-06-09
Context
256k
Max output
64,000
Parameters
30B (MoE, 3B active)
Architecture
Decoder-only Transformer, sparse mixture-of-experts (128 experts, 8 active per token), interleaved sliding-window attention with RoPE and global attention in 3:1 ratio, SwiGLU FFN with sigmoid-gated routing
Specialization
code
Openness
Open source
License
Apache 2.0(OSI)Commercial use allowed
Training
Two-stage: cascaded supervised fine-tuning (SFT) followed by reinforcement learning with verifiable rewards (RLVR), focusing on agentic coding
Created by

Empowering developers with advanced language AI.

Toronto, Ontario, Canada
Founded 2022
Website
Pricing
Output / 1M
-
Input / 1M
-

Cheapest of 1 route · Cohere API

About

A 30B-parameter sparse mixture-of-experts (3B active parameters) code generation model from Cohere, released as open weights under Apache 2.0. Designed for agentic software engineering, code generation, and terminal-based tasks. Supports a 256K token context window with 64K maximum output length, interleaved thinking, and tool use via JSON schema chat templates. The inaugural model in Cohere's North family.

North Mini Code 1.0 is an open-source model in the North family. The structured metadata tracks a 256k-token context window, reasoning, function calling, tool use, and structured outputs. This page tracks provider routes through Cohere API. Headline tracked benchmarks include SWE-bench Verified 61.0.

Top use-case fit: coding, agents, and build tasks

Coding

1 relevant benchmark in the decision map.

RAG

Included by capability and metadata signals in the decision map.

Agents

1 relevant benchmark in the decision map.

Capabilities

ReasoningFunction CallingTool UseStructured Outputs

Benchmark peer barsfor Coding

Benchmark scores(1)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
BenchmarkScoreVersionSource
SWE-bench Verified61.0Mini-SWE-Agent pass@1; Cohere self-reportedhttps://huggingface.co/blog/CohereLabs/introducing-north-mini-code

Migration checks

No linked migration route is available for this model yet.