Mi:dm K 2.5 Pro
Mi:dm K 2.5 Pro is a model from Korea Telecom with a 131k-token context window. No provider we track lists it yet.
Providers serving Mi:dm K 2.5 Pro
No provider we track lists this model yet.
Mi:dm K 2.5 Pro benchmark scores
Compare every model in Bench rankingsPublished scores only, each linked to its source. Ticks are every other model's best published score.
GPQA Diamond
PhD-level scientific reasoning, pass@1, no tools
70.1%, source: Artificial AnalysisRanks 148th of 221 models with a score
Humanity's Last Exam (text-only)
Reasoning & knowledge, no tools
8.1%, source: Artificial AnalysisRanks 149th of 221 models with a score
τ²-Bench Telecom
Agentic tool use (telecom), with a Qwen3 model as the simulated user
86.5%, source: Artificial AnalysisRanks 47th of 160 models with a score
LiveCodeBench
Competitive programming, Artificial Analysis's problem set (dates not published)
65.6%, source: Artificial AnalysisRanks 36th of 82 models with a score
IFBench
Instruction following, 294 prompts, 5 runs each
49.3%, source: Artificial AnalysisJoint 83rd of 158 models with a score
Terminal-Bench Hard
Agentic coding & terminal use, Artificial Analysis's 44-task hard subset
2.3%, source: Artificial AnalysisJoint 115th of 157 models with a score
LCR
Long-context reasoning, AA-LCR v1.1
9%, source: Artificial AnalysisJoint 180th of 211 models with a score
Omni Accuracy
Omniscience knowledge accuracy
17.9%, source: Artificial AnalysisRanks 132nd of 216 models with a score
Non-Hallucination
1 − hallucination rate
6.1%, source: Artificial AnalysisRanks 199th of 216 models with a score
Show as table
| Benchmark | Score |
|---|---|
GPQA Diamond PhD-level scientific reasoning, pass@1, no tools | 70.1%, source: Artificial Analysis |
Humanity's Last Exam (text-only) Reasoning & knowledge, no tools | 8.1%, source: Artificial Analysis |
τ²-Bench Telecom Agentic tool use (telecom), with a Qwen3 model as the simulated user | 86.5%, source: Artificial Analysis |
LiveCodeBench Competitive programming, Artificial Analysis's problem set (dates not published) | 65.6%, source: Artificial Analysis |
IFBench Instruction following, 294 prompts, 5 runs each | 49.3%, source: Artificial Analysis |
Terminal-Bench Hard Agentic coding & terminal use, Artificial Analysis's 44-task hard subset | 2.3%, source: Artificial Analysis |
LCR Long-context reasoning, AA-LCR v1.1 | 9%, source: Artificial Analysis |
Omni Accuracy Omniscience knowledge accuracy | 17.9%, source: Artificial Analysis |
Non-Hallucination 1 − hallucination rate | 6.1%, source: Artificial Analysis |