VerdictPal
Join

Models / Cohere

North Mini Code

North · Released 2026-06-09

Cohere's small coding-focused MoE model, covered and evaluated by Artificial Analysis on 9 Jun. We track it as a code-specialist candidate rather than a general frontier model.

Focused coding model for low-cost code edit, completion, and repo-assistant experiments where a full frontier model is wasteful.

#40 of 51 on AA Index · current snapshot
20 AA Index · 10 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite35
Context128K
Input / 1M—
Output / 1M—
Knowledge cutoffopen
Output speed96 t/s
TTFT (default API)0.23s
Statuslightweight
ReasoningCodingAgenticLong context
Reasoning
76
Coding
38
Agentic
36
Long context
32

46average across 4 tested families

Benchmark placements

Where North Mini Code places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.

BenchmarkFamilyRankScoreSourceDate
AA output speedPerformance#27/ 69Artificial Analysis2026-07-21
IFBenchReasoning#43/ 5758Artificial Analysis2026-07-21
SciCodeCoding#54/ 6538Artificial Analysis2026-07-21
GPQA DiamondReasoning#55/ 6976Artificial Analysis2026-07-21
Artificial Analysis Intelligence IndexReasoning#56/ 6720Artificial Analysis2026-07-21
Humanity's Last ExamReasoning#56/ 6510Artificial Analysis2026-07-21
Terminal-BenchAgentic#56/ 6936Artificial Analysis2026-07-21
AA-LCRLong context#58/ 6432Artificial Analysis2026-07-21
AA time to first tokenPerformance#58/ 69Artificial Analysis2026-07-21
τ³-Bench BankingAgentic#60/ 646Artificial Analysis2026-07-21
Family context

The three highest-scoring carded models in each capability family. Where North Mini Code shows up, it's highlighted.

ReasoningNorth Mini Code · 76
  1. 1
    Claude Mythos Preview
    95
  2. 2
    GPT-5.6 Sol
    94
  3. 3
    Kimi K3
    94
CodingNorth Mini Code · 38
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticNorth Mini Code · 36
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextNorth Mini Code · 32
  1. 1
    Kimi K3
    75
  2. 2
    GPT-5.1
    75
  3. 3
    GPT-5.6 Sol
    74
Newest receipts

Benchmark rows added to the public ledger for North Mini Code in the last 120 days. Older rows live in the full table below.

10 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context32.33%Artificial Analysis2
AA output speedPerformance96 t/sArtificial Analysis2
AA time to first tokenPerformance0.23sArtificial Analysis2
Artificial Analysis Intelligence IndexReasoning19.8%Artificial Analysis2
GPQA DiamondReasoning75.7%Artificial Analysis2
Humanity's Last ExamReasoning10%Artificial Analysis2
IFBenchReasoning57.55%Artificial Analysis2
SciCodeCoding38.2%Artificial Analysis2
τ³-Bench BankingAgentic6.39%Artificial Analysis2
Terminal-BenchAgentic35.58%Artificial Analysis2
Family coverage

A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.

Reasoning76
Coding38
Agentic36
Long context32
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover or tab any dot for name, score, and input price.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMiniMaxxAIMetaZhipuXiaomiMeituanCohere

Full benchmark ledger

Every catalog benchmark for North Mini Code. Scores link to the original source; gaps mean no public row exists yet.

10 sourced rows

10 of 32 catalog benchmarks have a sourced row for North Mini Code.

Reasoning

Coding

Agentic

Long context

Performance

Sources