A.X-K1: benchmark scores, price and ranking

A.X-K1 from skt ranks #130 of 743 LLMs on the LLMs Tiger aggregate benchmark score (51.3), based on 2 benchmark results captured up to 2026-10-13.

Organisation: skt
Type: Open-weight / local model
Open weights: Yes
Aggregate rank: #130 of 743 (score 51.3)

Across the 2 benchmarks where at least ten models are scored, A.X-K1 is in the top 10% on 0 and in the top half on 2.

A.X-K1 benchmark results

A.X-K1 benchmark scores, ranked against all models measured on each benchmark
Benchmark	Score	Rank	Source	Captured
MMLU-Pro	81.5%	#28 of 136	HuggingFace MMLU-Pro leaderboard	2026-10-13
GPQA Diamond	74.0%	#133 of 275	HuggingFace GPQA Diamond leaderboard	2026-10-13

About A.X-K1

A.X-K1 is an open-weight model that can run on your own hardware from skt. No API price is listed for it yet. We hold 2 benchmark results for it across 1 area, and none is in the top tenth of its board.

Where A.X-K1 stands

On our aggregate score, A.X-K1 is #130 of 743 ranked models, in the top quarter. Its score is 51.3; the scale is centred on 50, so it sits 1.3 points above the typical model. That beats 83% of the ranked field. Among the 2 ranked skt models it is #2. The next one up is A.X-K2 at 51.4.

A.X-K1, benchmark by benchmark

MMLU-Pro puts A.X-K1 at 81.5%, #28 of 136. That beats 79% of the models tested, but 6.6 points short of Darwin-180B-RSI (88.1%). The results either side are NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4 (81.6%) and GLM-4.5-Air (81.4%). It covers 57-subject knowledge + reasoning. The board median is 55.3%, so A.X-K1 is 26.2 points above it. Source: HuggingFace MMLU-Pro leaderboard, captured 2026-10-13.

On GPQA Diamond, A.X-K1 reaches 74.0%. It ranks #133 of 275, ahead of 52% of the field, and 21.8 points behind GPT-6 Astra (max) at 95.8%. The results either side are JoyAI-LLM-Flash (74.4%) and claude-opus-4-1-20250805 (73.2%). It covers expert-level science questions. The board median is 71.7%, so A.X-K1 is 2.3 points above it. Source: HuggingFace GPQA Diamond leaderboard, captured 2026-10-13.

Strengths and weaknesses

Reasoning, maths and science: solid, ahead of 66% of the field on average across 2 benchmarks. Its best board here is MMLU-Pro (#28 of 136) and its weakest is GPQA Diamond (#133 of 275).

Reading these numbers

The 2 scores on this page come from 2 sources: HuggingFace GPQA Diamond leaderboard (1), HuggingFace MMLU-Pro leaderboard (1). No official leaderboard supplied any of them, so treat them as third-party measurements. With 2 results the aggregate rank can move a lot when one new score arrives.

Models ranked nearby

#127 DeepSeek Chat V2 0628 (51.4)
#128 MiniMax-M1-40k (51.4)
#129 Darwin-4B-David (51.3)
#131 JoyAI-LLM-Flash (51.3)
#132 Claude Opus 4.6 (120K, Medium) (51.3)
#133 seed-oss-36b-instruct (51.3)