Benchmarks of MiniCPM5 2B16
BenchmarkByScoreRank
Overall1
Knowledge3
Science3
Coding2
Agents4
Documents2
Legacy · saturated or no longer reported1
No benchmark matches
—per task?
price not verifiedprice
—output speed?
131Kcontext
Rankingin thinking mode?
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGPT-6 Lunalow+73%—$0.005Function calling only with none · Chat Completions+3IOpenK2 Horizon 7Bthinking+65%—$0.30 / $1.20Memory (4-bit): 1.3 GB → 4.5 GBBetterGPT-5.6 Lunalow+69%—$0.01
BetterClaude Haiku 5.5low×2.4—$0.024Remove sampling parameters+5
BetterLing 3.0 FlashthinkingOpen+62%—$0.028Memory (4-bit): 1.3 GB → 76 GBModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
GPT-6 Lunalow+73%—$0.005Function calling only with none · Chat Completions+3GPT-5.6 Lunalow+69%—$0.01Claude Haiku 5.5low×2.4—$0.024Remove sampling parameters+5Ling 3.0 FlashthinkingOpen+62%—$0.028Memory (4-bit): 1.3 GB → 76 GB
MiMo v2.5thinkingOpen+75%—$0.042Memory (4-bit): 1.3 GB → 186 GBModelScoreCostPer taskWhat changes
IK2 Horizon 7BthinkingOpen+65%—$0.30 / $1.20Memory (4-bit): 1.3 GB → 4.5 GBIK2 Horizon 3.7BthinkingOpen+25%—$0.30 / $1.20Memory (4-bit): 1.3 GB → 2.5 GB
Gemma 4 12BthinkingOpen+14%—$0.10 / $0.30Memory (4-bit): 1.3 GB → 6 GB
Qwen 3.5 9Bnon-thinkingOpensame—$0.08 / $0.13Memory (4-bit): 1.3 GB → 4.8 GBQwen3.5 4BthinkingOpensame—$0.03 / $0.15Memory (4-bit): 1.3 GB → 2.3 GBRun it yourself
estimateFits on✓ 8 GB (phone, laptop) · full quality✓ 16 GB · full quality✓ 24 GB GPU · full quality✓ 48 GB GPU · full quality✓ 80 GB GPU · full quality
LicenceApache-2.0free for commercial use
Benchmarks
all categoriesChat & Knowledge
Humanity's Last Exam8.9#206
AA Omniscience Index-11.7#104
CritPt (physics)0.3#151
AA Intelligence Index 4.3.212.5#179
AA Omniscience · Accuracy8.4#296
AA-LCR (long context) 1.159.3#163
GDP.pdf · AA all-pass1.4#114
GPQA Diamond70.2#183
Coding
SciCode26.3#129
Terminal-Bench 2.18.6#129
Agents & Tool Use
AutomationBench-AA1.8#103
GDPval-AA (work tasks) 2.1707.2#93
τ³-Banking · AA20.8#53
AA-Briefcase 1.1 · Elo422.6#103
FAQ
How much does MiniCPM5 2B cost?›
There is no per-unit API price: price not verified.
How good is MiniCPM5 2B, and what is it best at?›
#205 of 336 in Chat & Knowledge, #77 of 134 in Coding, #103 of 135 in Agents & Tool Use, #114 of 135 in Vision & Documents.
Is MiniCPM5 2B better than Qwen3.5 Omni Flash?›
On AA Intelligence Index 4.3.2, MiniCPM5 2B scores 12.5 vs 12.5 — about the same.
What is the context window of MiniCPM5 2B?›
131K tokens.
When was MiniCPM5 2B released? Is there a newer version?›
Released 2026-09-07. It is the current version in its line.
Is MiniCPM5 2B open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run MiniCPM5 2B locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can MiniCPM5 2B read images?›
No, it accepts text only.
What is MiniCPM5 2B?›
MiniCPM5 2B is an open-weights language model from OpenBMB, released Sep 2026.
What are the best alternatives to MiniCPM5 2B?›
GPT-6 Luna (better), K2 Horizon 7B (open weights).