Ring-flash-2.0
Ring-flash-2.0 is an open-weights model from InclusionAI with a 32.8k-token context window. No provider we track lists it yet.
Providers serving Ring-flash-2.0
No provider we track lists this model yet.
Ring-flash-2.0 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
72.5%, source: Artificial AnalysisRanks 145th of 221 models with a score
Humanity's Last Exam (text-only)
Reasoning & knowledge, no tools
9.6%, source: Artificial AnalysisRanks 144th of 221 models with a score
τ²-Bench Telecom
Agentic tool use (telecom), with a Qwen3 model as the simulated user
0%, source: Artificial AnalysisJoint 155th of 160 models with a score
LiveCodeBench
Competitive programming, Artificial Analysis's problem set (dates not published)
62.8%, source: Artificial AnalysisRanks 40th of 82 models with a score
IFBench
Instruction following, 294 prompts, 5 runs each
43.3%, source: Artificial AnalysisRanks 96th of 158 models with a score
Terminal-Bench Hard
Agentic coding & terminal use, Artificial Analysis's 44-task hard subset
7.6%, source: Artificial AnalysisJoint 89th of 157 models with a score
LCR
Long-context reasoning, AA-LCR v1.1
21%, source: Artificial AnalysisJoint 163rd of 211 models with a score
Omni Accuracy
Omniscience knowledge accuracy
16.6%, source: Artificial AnalysisRanks 143rd of 216 models with a score
Non-Hallucination
1 − hallucination rate
9.9%, source: Artificial AnalysisRanks 174th of 216 models with a score
Show as table
| Benchmark | Score |
|---|---|
GPQA Diamond PhD-level scientific reasoning, pass@1, no tools | 72.5%, source: Artificial Analysis |
Humanity's Last Exam (text-only) Reasoning & knowledge, no tools | 9.6%, source: Artificial Analysis |
τ²-Bench Telecom Agentic tool use (telecom), with a Qwen3 model as the simulated user | 0%, source: Artificial Analysis |
LiveCodeBench Competitive programming, Artificial Analysis's problem set (dates not published) | 62.8%, source: Artificial Analysis |
IFBench Instruction following, 294 prompts, 5 runs each | 43.3%, source: Artificial Analysis |
Terminal-Bench Hard Agentic coding & terminal use, Artificial Analysis's 44-task hard subset | 7.6%, source: Artificial Analysis |
LCR Long-context reasoning, AA-LCR v1.1 | 21%, source: Artificial Analysis |
Omni Accuracy Omniscience knowledge accuracy | 16.6%, source: Artificial Analysis |
Non-Hallucination 1 − hallucination rate | 9.9%, source: Artificial Analysis |