A.X K2 measured on 9 of 104 Chat & Knowledge benchmarks9
BenchmarkByScoreRank
Overall1
Knowledge3
Science2
Documents2
Legacy · saturated or no longer reported1
No benchmark matches
—per task?
self-host onlyprice
—output speed?
262Kcontext
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGPT-6 Lunahigh+11%—$0.029Function calling only with none · Chat Completions+3
OpenMiMo v2.6 Prothinking+67%—$0.13Memory (4-bit): 415 GB → 615 GB+4BetterGPT-5.6 Lunahigh+13%—$0.044
BetterClaude Haiku 5.5medium+14%—$0.047Remove sampling parameters+5BetterMiMo v2.5 ProthinkingOpen+21%—$0.054Memory (4-bit): 415 GB → 614 GBModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
GPT-6 Lunahigh+11%—$0.029Function calling only with none · Chat Completions+3GPT-5.6 Lunahigh+13%—$0.044Claude Haiku 5.5medium+14%—$0.047Remove sampling parameters+5MiMo v2.5 ProthinkingOpen+21%—$0.054Memory (4-bit): 415 GB → 614 GBMiMo v2.6 FlashthinkingOpen+19%—$0.062Pass back reasoning_content+3ModelScoreCostPer taskWhat changes
MiMo v2.6 ProthinkingOpen+67%—$0.13Memory (4-bit): 415 GB → 615 GB+4
Step 5 PreviewthinkingOpen+57%—$1.03web_search not available · Chat Completions+2
Muse Spark OpenxhighOpen+48%—$0.84Memory (4-bit): 415 GB → 493 GB+2
DeepSeek V4.1 Flash 1017maxOpen+47%—$0.26Memory (4-bit): 415 GB → 458 GB+5
GLM 5.3maxOpen+43%—$2.01Memory (4-bit): 415 GB → 452 GB+4Run it yourself
estimateFits on✗ 8 GB (phone, laptop)✗ 16 GB✗ 24 GB GPU✗ 48 GB GPU✗ 80 GB GPU
LicenceApache-2.0free for commercial use
Benchmarks
in Chat & KnowledgeAA Intelligence Index 4.3.221.5#109
AA Omniscience Index-8.2#88
AA Omniscience · Accuracy18.6#172
Humanity's Last Exam29.6#83
GPQA Diamond85.7#71
CritPt (physics)8.9#65
GDP.pdf · AA all-pass6.4#76
FAQ
How much does A.X K2 cost?›
There is no per-unit API price: self-host only.
Is A.X K2 free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is A.X K2, and what is it best at?›
#83 of 334 in Chat & Knowledge, #75 of 132 in Coding, #94 of 133 in Agents & Tool Use, #76 of 133 in Vision & Documents. Strongest at: Reasoning, Science.
Is A.X K2 better than MiniMax M2.7?›
On Humanity's Last Exam, A.X K2 scores 29.6 vs 29.6 — about the same.
What is the context window of A.X K2?›
262K tokens.
When was A.X K2 released? Is there a newer version?›
Released 2026-08-12. It is the current version in its line.
Is A.X K2 open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run A.X K2 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can A.X K2 read images?›
No, it accepts text only.
What is A.X K2?›
A.X K2 is an open-weights language model from SK Telecom, released Aug 2026.
What are the best alternatives to A.X K2?›
GPT-6 Luna (better), MiMo v2.6 Pro (open weights).