The LLM benchmark aggregator.

K2 Horizon 0.9B

Institute of Foundation Models · 0.9B

Strongest in Knowledge work (#178 of 178, on 1 of its 2 boards), weakest in Agents (#266 of 268). Above par in 3 of 10 scopes. Among the models it meets almost everywhere, it finishes behind Claude Sonnet 5 and Claude Fable 5 and ahead of GPT-4o Mini and Qwen3.6 Plus.

§ 1 · Profile

What it is good at

Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.

Core abilities40.4−27#186/2041/3

General intelligence36.1−34#193/2041/3

Agents37−31.1#266/2681/5

Knowledge work30.5−43.1#178/1781/2

Tool use38.8−35.2—/81✱0/1

Reasoning49.1−17.8—/178✱0/4

Mathematics51.3−14.4—/140✱0/2

Science32.1−31.4—/122✱0/1

Coding52.1−17—/165✱0/5

Code generation55.5−14.5—/77✱0/2

§ 2 · Head to head

What it beats, and what beats it

The same models turn up scope after scope. Counted once: where both were placed, who finished higher. Bars run right for K2 Horizon 0.9B, left for the other.

OpenBMB 1Bwon 6 · lost 1

Ox Alphawon 5 · lost 1

Qwen3.5 0.8Bwon 6 · lost 2

Qwen3.6 27Bwon 6 · lost 2

Claude 3.5 Haikuwon 4 · lost 2

Grok 4.3won 6 · lost 3

Kimi K2.7 Codewon 5 · lost 3

Nemotron 3 Ultra 550B A55Bwon 5 · lost 3

G9v3 3Bwon 4 · lost 3

Gemini 3.5 Flash Litewon 5 · lost 4

GPT-5.4 Miniwon 5 · lost 4

Qwen3.6 Pluswon 5 · lost 4

GPT-4o Miniwon 5 · lost 4

Gemma 4 E4Bwon 3 · lost 3

Inkling Smallwon 3 · lost 3

Mimo V2.5won 3 · lost 3

Gemini 2.0 Flashwon 3 · lost 3

GPT-5.2 Codexwon 3 · lost 3

NVIDIA Nemotron 3 Nano 4Bwon 4 · lost 4

Grok Build 0.1won 4 · lost 4

Command Awon 4 · lost 4

Mistral Medium 3.5won 4 · lost 4

Kimi K2.6won 4 · lost 5

DeepSeek V3won 4 · lost 5

Mercury 2.5won 3 · lost 4

G9v3 39A5Bwon 3 · lost 4

Claude Opus 4won 3 · lost 4

MiniMax M3won 4 · lost 6

Inklingwon 4 · lost 6

Qwen3.5 2Bwon 3 · lost 5

Granite 4.2 30Bwon 3 · lost 5

Mistral Large 3won 3 · lost 5

Nemotron 3 Superwon 3 · lost 5

Llama 4 Scout Instructwon 3 · lost 5

GPT-4.1won 3 · lost 5

Mimo V2.5 Prowon 3 · lost 5

Qwen3.8 Flash Nextwon 3 · lost 5

DeepSeek V4 Flash Visionwon 3 · lost 5

GPT-5.4 Nanowon 3 · lost 6

GPT-4.1 Nanowon 3 · lost 6

Qwen3.6 35B A3Bwon 3 · lost 6

GLM-5.3 Flashwon 3 · lost 6

GPT-4.1 Miniwon 3 · lost 6

Nemotron 3 Ultrawon 3 · lost 6

Gemini 3.6 Flashwon 3 · lost 6

Llama 4 Maverick Instructwon 3 · lost 6

Qwen3.7 Maxwon 3 · lost 6

Gemini 2.5 Flashwon 3 · lost 6

Gemini 3.5 Flashwon 3 · lost 6

Grok 4.5won 3 · lost 6

Granite 4.2 3Bwon 3 · lost 7

Qwen3.8 27Bwon 3 · lost 7

Qwen3.5 4Bwon 2 · lost 5

Mistral Smallwon 2 · lost 5

Gemini 2.5 Flash Litewon 2 · lost 5

Gemma 4 12Bwon 2 · lost 5

K2 Horizon 32Bwon 2 · lost 5

Claude Opus 4.6won 2 · lost 5

Muse Glimmer-30Bwon 2 · lost 5

GPT OSS 120Bwon 2 · lost 6

Qwen3.5 9Bwon 2 · lost 6

Kimi K2won 2 · lost 6

Muse Spark 1.2won 2 · lost 6

Granite 4.2 8Bwon 2 · lost 7

Claude Sonnet 4.6won 2 · lost 7

K2 Horizon 7Bwon 2 · lost 7

Muse Spark 1.1won 2 · lost 7

K2 Horizon MoVA 36B A4Bwon 2 · lost 7

DeepSeek V4 Flashwon 2 · lost 7

Claude Opus 4.5won 2 · lost 7

GPT-5.2won 2 · lost 7

GLM-5.3won 2 · lost 7

Gemma 4 31Bwon 2 · lost 7

Qwen3 235B A22B Instructwon 2 · lost 7

Gemini 3.8 Flashwon 2 · lost 7

Claude Sonnet 4won 2 · lost 7

Gemini 3.1 Prowon 2 · lost 7

K2 Horizon 3.7Bwon 2 · lost 8

GLM-5.2won 2 · lost 8

Muse Sparkwon 1 · lost 5

MiniMax M2.7won 1 · lost 5

GPT-5.1won 1 · lost 5

GLM-4.7won 1 · lost 5

GLM-5.1won 1 · lost 5

Grok 4.1 Fastwon 1 · lost 5

Gemma 3 27B Itwon 1 · lost 5

Qwen3 32Bwon 1 · lost 5

MiniMax M2.5won 1 · lost 5

Kimi K2 Thinkingwon 1 · lost 5

Grok 4 Fastwon 1 · lost 5

Qwen3 Maxwon 1 · lost 5

GLM-5won 1 · lost 5

Grok 4.20won 1 · lost 5

O3 Miniwon 1 · lost 6

GPT-5 Nanowon 1 · lost 6

GPT-6 Lunawon 1 · lost 7

Gemini 3.1 Flash Litewon 1 · lost 7

DeepSeek R1won 1 · lost 7

Grok 4won 1 · lost 7

Grok 4.7won 1 · lost 7

Kimi K2.5won 1 · lost 7

Gemini 3 Prowon 1 · lost 7

Claude Haiku 4.5won 1 · lost 8

GLM-4.6won 1 · lost 8

DeepSeek V3.2won 1 · lost 8

Claude Sonnet 4.5won 1 · lost 8

K2 Horizon 375B A23Bwon 1 · lost 8

DeepSeek V4 Prowon 1 · lost 8

O4 Miniwon 1 · lost 8

Gemini 2.5 Prowon 1 · lost 8

GPT-5 Miniwon 1 · lost 8

GPT-5.4won 1 · lost 8

Gemini 3.7 Flashwon 1 · lost 8

Grok 4.6won 1 · lost 8

DeepSeek V4.1 Flashwon 1 · lost 8

Qwen3.8 Maxwon 1 · lost 8

GPT-5.6 Lunawon 1 · lost 9

GPT-5.6 Terrawon 1 · lost 9

Union Alphawon 0 · lost 6

GPT OSS 20Bwon 0 · lost 7

Grok 3 Miniwon 0 · lost 7

Mimo V2.6 Prowon 0 · lost 7

GPT-5won 0 · lost 8

GPT-6 Solwon 0 · lost 8

Claude Opus 5.5won 0 · lost 8

Muse Spark 1.3won 0 · lost 8

GPT-6 Astrawon 0 · lost 8

Claude Opus 4.8won 0 · lost 9

O3won 0 · lost 9

Claude Opus 4.7won 0 · lost 9

Claude Opus 5won 0 · lost 9

Kimi K3won 0 · lost 9

GPT-5.6 Solwon 0 · lost 9

GPT-5.5won 0 · lost 9

Claude Fable 5.1won 0 · lost 9

Claude Fable 5won 0 · lost 9

Claude Sonnet 5won 0 · lost 10

§ 3 · Sources

Where the numbers come from

3 publications, 10 figures. Every one links to the page it was read from.

Artificial Analysis ↗1 measureread 2026-09-26

Artificial Analysis Intelligence Index 3

GDPval-AA ↗1 measureread 2026-09-26

GDPval-AA -426

✱✱ IFM — K2 Horizon launch ↗the model’s own publisher8 measuresread 2026-09-26

GPQA Diamond 27.3%HMMT Feb 2026 25.8%BFCL v4 28%HumanEval+ 79.9%MBPP+ 68%AIME 2025 41.7%AIME 2026 48.5%LiveCodeBench v6 37.4%

Badge

[![PublicAI Index](https://publicai.io/model-index/badge?model=k2-horizon-0-9b)](https://publicai.io/model-index/m/k2-horizon-0-9b)