Benchmarks of K2 Think V214
BenchmarkByScoreRank
Overall1
Knowledge4
Science2
Coding1
Agents2
Documents1
Legacy · saturated or no longer reported3
No benchmark matches
—per task?
price not verifiedprice
—output speed?
262Kcontext
Ranking?
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGPT-6 Lunalow+87%—$0.005Function calling only with none · Chat Completions+3
OpenQwen3.8 Flashthinking×3.5—$0.37Memory (4-bit): 44 GB → 108 GB+6BetterGPT-5.6 Lunalow+83%—$0.01
BetterMiMo v2.5thinkingOpen+90%—$0.019Memory (4-bit): 44 GB → 186 GB
BetterClaude Haiku 5.5low×2.6—$0.024Remove sampling parameters+5ModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
GPT-6 Lunalow+87%—$0.005Function calling only with none · Chat Completions+3GPT-5.6 Lunalow+83%—$0.01MiMo v2.5thinkingOpen+90%—$0.019Memory (4-bit): 44 GB → 186 GBClaude Haiku 5.5low×2.6—$0.024Remove sampling parameters+5GPT-5 Minihigh+46%—$0.053ModelScoreCostPer taskWhat changes
Qwen3.8 FlashthinkingOpen×3.5—$0.37Memory (4-bit): 44 GB → 108 GB+6Qwen3.8 27BxhighOpen×2.9—$1.01max_tokens range · QwenCloud+4
Nemotron 3.5 Super VLOpen×2.3—$0.19Memory (4-bit): 44 GB → 75 GBIK2 Horizon MoVA 36B A4BthinkingOpen×2.2—$1 / $4
Ling 3.0 Flash VLthinkingOpen×2.1—$0.06 / $0.18Memory (4-bit): 44 GB → 75 GBModelScoreCostPer taskWhat changes
IK2 Horizon 375B A23BOpen×2.7—$1 / $4Memory (4-bit): 44 GB → 228 GB+1IK2 Horizon MoVA 36B A4BthinkingOpen×2.2—$1 / $4IK2 Horizon 7BthinkingOpen+79%—$0.30 / $1.20IK2 Horizon 3.7BthinkingOpen+36%—$0.30 / $1.20IK2 v2highOpen-14%——Memory (4-bit): not verifiedIK2 Horizon 0.9BthinkingOpen-73%—$0.30 / $1.20Context: 262K → 131KAll Institute of Foundation Models models →Run it yourself
estimateFits on✗ 8 GB (phone, laptop)✗ 16 GB✗ 24 GB GPU✓ 48 GB GPU · compressed✓ 80 GB GPU · 8-bit
LicenceApache-2.0free for commercial use
Benchmarks
all categoriesChat & Knowledge
Humanity's Last Exam10.1#192
AA Omniscience Index-40.3#175
CritPt (physics)0#180
AA Intelligence Index 4.3.2≈11.5#189
GPQA Diamond71.3#180
Coding
Terminal-Bench 2.115#118
Agents & Tool Use
τ²-bench · AA25.4#172
Writing & Language
IFBench (instructions)62.8#79
FAQ
How much does K2 Think V2 cost?›
There is no per-unit API price: price not verified.
How good is K2 Think V2, and what is it best at?›
#191 of 334 in Chat & Knowledge, #142 of 239 in Coding, #122 of 175 in Agents & Tool Use, #79 of 245 in Writing & Language.
Is K2 Think V2 better than GPT OSS 120B?›
On AA Intelligence Index 4.3.2, K2 Think V2 scores 11.5 vs 11.6 — about the same.
What is the context window of K2 Think V2?›
262K tokens.
When was K2 Think V2 released? Is there a newer version?›
Released 2025-12-15. It is the current version in its line.
Is K2 Think V2 open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run K2 Think V2 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can K2 Think V2 read images?›
No, it accepts text only.
What is K2 Think V2?›
K2 Think V2 is an open-weights language model from Institute of Foundation Models, released Dec 2025.
What are the best alternatives to K2 Think V2?›
GPT-6 Luna (better), Qwen3.8 Flash (open weights).