Inference
EN
Cohere

North Mini Code

North Mini Code is a reasoning model from Cohere. 9 benchmarks count toward its index, in 5 categories.

  • Open weights
  • Reasoning
  • Input: text
  • Released 2026-06
RankedEnough results for an index.
IndexAll categories, out of 100.38.0±7.9Median 51.0
CoverageIndex weight that has results.75%
SpeedOutput tokens per second.70/sMedian 67/s
First tokenSeconds to the first output token.28.9 sMedian 1.3 s
Input / 1MUS dollars per 1M input tokens.FreeMedian $0.60
Output / 1MUS dollars per 1M output tokens.FreeMedian $2.55
ContextMaximum tokens in one request.256KMedian 262K
EloLMArena rating and rank.—Median 1422

Run locally

30.48B parameters
16-bit
65.0 GB
8-bit
35.0 GB
4-bit
20.4 GB
Weights
Hugging Face (opens in a new tab) ↗

Memory need with a short context. The 4-bit copy fits in 24 GB. Compare local models

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.2 counted30.8198 / 218
CodingCode writing and repair.1 counted46.4—
ReasoningLogic problems and puzzles.2 counted35.2190 / 227
MultimodalTasks with images and text.None counted——
KnowledgeFacts and expert knowledge.3 counted38.7203 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.1 counted42.9—
MathMath problems.None counted——

Results

9 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
Artificial Analysis IFBenchInstruction57.6%66 / 146
Artificial Analysis GPQA DiamondKnowledge75.7%114 / 199
Artificial Analysis Humanity's Last ExamKnowledge11.1%140 / 222
Critical Physics TasksReasoning0.3%137 / 214
τ²-Bench Tool-Agent-User EvaluationAgentic37.4%105 / 149
GDPval-AA normalizedAgentic0.0%101 / 133
Artificial Analysis Long Context ReasoningReasoning37.3%172 / 217
Artificial Analysis SciCodeCoding38.8%93 / 114
Artificial Analysis Omniscience AccuracyKnowledge18.4%137 / 164

Sources

BenchLMCC BY-NC 4.0

History

5 releases
Index38.6 → 38.0
23 Oct 202612 Nov 2026
Output / 1MFree → Free
23 Oct 202612 Nov 2026

Questions

How much does North Mini Code cost?
Input tokens are free, output tokens are free.
How good is North Mini Code?
Its index is 38.0 out of 100. That is place 220 of 283 ranked models.
What is North Mini Code best at?
Knowledge: 38.7 out of 100, place 203 of 268.
How fast is North Mini Code?
70 output tokens per second, 28.9 s to the first token.
What is the context window of North Mini Code?
256K tokens.
Is North Mini Code open source?
Yes. Its weights are published.
When was North Mini Code released?
June 2026.

Badge

North Mini Code on Inference360: place 220 of 283, index 38.0
HTML
<a href="https://inference360.ai/models/north-mini-code-1-0"><img src="https://inference360.ai/badges/north-mini-code-1-0.svg" alt="North Mini Code on Inference360: place 220 of 283, index 38.0" height="20"></a>
Markdown
[![North Mini Code on Inference360: place 220 of 283, index 38.0](https://inference360.ai/badges/north-mini-code-1-0.svg)](https://inference360.ai/models/north-mini-code-1-0)

Compare with

GPT-4.138.0DeepSeek V3.137.9Mistral Small 437.9Command A+40.2

More from Cohere

All models
Command A+40.2North Micro Vision InstructUnrankedNorth Small TranslateUnrankedCommand R7B (12-2024)UnrankedCommand R (08-2024)Unranked