As of: 2026-10-25 23:49 CET.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/m/k2-horizon-375b-a23b/

# K2 Horizon 375B-A23B

MBZUAI Institute of Foundation Models (IFM), K2 Horizon family. Status: current. Type: reasoning model. Released: 2026-09-03. Training cutoff: not published.

## How to call it

| Provider | Model id | Endpoint | Docs |
|---|---|---|---|
| Hugging Face (open weights, BF16) | `IFM/K2-Horizon-375B-A23B` |  | https://huggingface.co/IFM/K2-Horizon-375B-A23B |
| Hugging Face (FP8) | `IFM/K2-Horizon-375B-A23B-FP8` |  | https://huggingface.co/IFM/K2-Horizon-375B-A23B-FP8 |
| Featherless | `IFM/K2-Horizon-375B-A23B` |  | https://featherless.ai/models/IFM/K2-Horizon-375B-A23B |
| Vultr (GPU deployment guide) |  |  | https://docs.vultr.com/models/ifm-k2-horizon/k2-horizon-375b-a23b |

## Pricing

- Price not published.
- Last checked: 2026-10-25

## Specs

- Input: text
- Output: text
- Context window: 524,288 tokens
- Max output: not published
- Open weights: yes
- Licence: apache-2.0

## What it has not seen

- No published training cutoff. Counted from its release (2026-09-03): at least 1,056 AI events since (212 major, 27 historic), as of 2026-10-25.
- Everything since August 2026: https://postcutoff.com/since/2026-08/

## What stands out

- Fully open frontier-scale MoE: 375B total / 23B active sparse MoE released with weights, code (github.com/ifm-ai/xllm), pretraining and midtraining data (TxT360-v2) and W&B training logs, all Apache 2.0; MBZUAI calls K2 Horizon the largest fully open model release. Source: https://huggingface.co/IFM/K2-Horizon-375B-A23B
- Native 512K context: 524,288-token context native from midtraining onward. Source: https://huggingface.co/IFM/K2-Horizon-375B-A23B
- Fastest model on Artificial Analysis (output speed, IFM API) (found after launch): AA measured 507.4 output tokens/s on IFM's API, #1 of 78 in its comparison set, with Intelligence Index 31 (AA v4.3.2), the top non-Chinese open-weights score after Meta's Muse Spark Open per AA's Oct 16 post. Source: https://artificialanalysis.ai/models/k2-horizon-375b-a23b
- Agentic benchmarks with an audited coding score: Model card: Toolathlon Verified 65.3, BrowseComp 72.8, Terminal-Bench 2.1 70.2, GPQA Diamond 87.3, HLE (no tools) 32.0 (developer-reported; third-party coverage says an IFM reward-hacking audit lowered Terminal-Bench 2.1 to 66.9, not verified here). Source: https://huggingface.co/IFM/K2-Horizon-375B-A23B

## Notes

Maker: MBZUAI's Institute of Foundation Models (IFM, Abu Dhabi; successor of the LLM360 K2 line, W&B project llm360). Family of six models: 0.9B, 3.7B, 7B, 32B dense, 36B-A4B MoE and this 375B-A23B flagship (MBZUAI news, Sept 2026). Release date Sept 3, 2026 per Artificial Analysis FAQ and press (one listing says Sept 1). Reasoning model: set reasoning_effort via chat_template_kwargs, thinking in reasoning_content; vLLM (--reasoning-parser k2_horizon --tool-call-parser k2_horizon, TP 8 + expert parallel) and SGLang (8x H200). Artificial Analysis (read 2026-10-25, https://artificialanalysis.ai/models/k2-horizon-375b-a23b): Intelligence Index 31, output speed 507.4 tok/s (#1 of 78), TTFAT 6.05 s, single API provider (IFM) shown at $0.00/$0.00 — the price looks like a placeholder or free access, so no pricing block; IFM's hosted API endpoint and model id were not verified (ifm.ai returned 403). Featherless listing seen in search, not opened. Base model for Jais 3 (MBZUAI/Inception42, Oct 15, 2026). No timeline entry for the K2 Horizon launch itself yet.

The largest fully open (weights + data + code + logs) model family as of its release, from the UAE's MBZUAI. Fast on AA because IFM serves it at high throughput; intelligence is mid-tier.

Sources: [Hugging Face model card](https://huggingface.co/IFM/K2-Horizon-375B-A23B) ·
[MBZUAI news](https://mbzuai.ac.ae/news/mbzuais-institute-of-foundation-models-launches-k2-horizon-the-worlds-largest-fully-open-ai-mo) ·
[IFM blog](https://ifm.ai/blog/k2/) · [Artificial Analysis](https://artificialanalysis.ai/models/k2-horizon-375b-a23b) ·
[Gulf Business](https://gulfbusiness.com/en/2026/artificial-intelligence/mbzuai-launches-k2-horizon-family-of-six-open-ai-models/)

Model page: https://postcutoff.com/m/k2-horizon-375b-a23b/