K2 Think V2
K2 Think V2 is behind the leaders in instruction following, long context, factuality, and coding. Too few results yet to rate reasoning, agentic tasks, safety, math, multimodal tasks, or multilingual tasks.
Capability profile
Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.
Too few results yet to rate Reasoning, Agentic, Safety, Math, Multimodal or Multilingual.
Price
No current price is tracked for this model. See the rate card
Evidence
18results on17benchmarks
- 17 aggregator
- 1 cross-referenced
From 2 sources · latest Oct 8, 2026 · How verification works
K2 Think V2 benchmark results
18 results on 17 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.
32.9% behind the leader1 of 3 ranked benchmarks measured
- IFBench62.79Oct 8, 2026aa_ifbench
33.0% behind the leader1 of 3 ranked benchmarks measured
- 57.00Oct 8, 2026
36.2% behind the leader2 of 4 ranked benchmarks measured
- AA-Omniscience · Non-hallucination28.83Oct 8, 2026omniscienceNonHallucination
- AA-Omniscience · Accuracy18.02Oct 8, 2026omniscienceAccuracy
43.9% behind the leader3 of 10 ranked benchmarks measured
- SciCode32.99Sep 4, 2026aa_scicode
- Terminal-Bench 2.114.98Oct 8, 2026terminalbenchV21
- Terminal-Bench Hard6.82Oct 8, 2026aa_terminalbench_hard
0 of 6 ranked benchmarks measured
- 0.00Oct 8, 2026
- GPQA Diamond71.31Oct 8, 2026gpqa
- Humanity's Last Exam10.10Oct 8, 2026aa_hle
0 of 7 ranked benchmarks measured
- 0.00Oct 8, 2026
- τ-Bench V3 · Banking5.36Aug 10, 2026tauBanking
More results
Benchmarks outside the capability baskets. They are not ranked against a leader.
- τ²-Bench Telecom (AA run)25.44Oct 8, 2026aa_tau2
- AA Intelligence11.50Oct 8, 2026aa_intelligence_index
- -40.33Oct 8, 2026
- Artificial Analysis Coding Index20.98Sep 9, 2026aa_coding_index
- 1.79Aug 10, 2026
- AA Intelligence11.00Sep 18, 2026Artificial Analysis Intelligence Index
K2 Think V2: common questions
Who makes K2 Think V2?
K2 Think V2 is made by MBZUAI.
When was K2 Think V2 released?
K2 Think V2 was released on Dec 15, 2025, according to Artificial Analysis.
What is K2 Think V2 good at?
K2 Think V2 is behind the leaders in instruction following, long context, factuality, and coding. Too few results yet to rate reasoning, agentic tasks, safety, math, multimodal tasks, or multilingual tasks.
How many benchmarks has K2 Think V2 been tested on?
We track 18 results for K2 Think V2 on 17 benchmarks from 2 sources. The latest was recorded on Oct 8, 2026.
About this record
Where K2 Think V2's numbers come from, and every name it appears under.
- Tracked since
- Jun 29, 2026
- Newest source mention
- Jun 29, 2026
Where the results come from
Verification: 18 scores · 0 independently verified · 17 aggregator-attributed · 1 vendor cross-reference. How these tiers are assigned
From 2 sources on 1 site. Artificial Analysis supplies all of them. Bars are coloured by trust tier.
- artificialanalysis.ai18