Benchmarks of MiniCPM5 1B11
BenchmarkByScoreRank
Overall1
Knowledge4
Science2
Coding1
Documents1
Legacy · saturated or no longer reported2
No benchmark matches
—per task?
self-host onlyprice
—output speed?
131Kcontext
Rankingin thinking mode?
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGPT-6 Lunalow×2.4—$0.005Function calling only with none · Chat Completions+3IOpenK2 Horizon 7Bthinking×2.3—$0.30 / $1.20Memory (4-bit): 0.5 GB → 4.5 GBBetterGPT-5.6 Lunalow×2.4—$0.01
BetterMistral Small 4thinkingOpen+28%—$0.015Memory (4-bit): 0.5 GB → 72 GB
BetterMiMo v2.5thinkingOpen×2.5—$0.019Memory (4-bit): 0.5 GB → 186 GBModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
GPT-6 Lunalow×2.4—$0.005Function calling only with none · Chat Completions+3GPT-5.6 Lunalow×2.4—$0.01Mistral Small 4thinkingOpen+28%—$0.015Memory (4-bit): 0.5 GB → 72 GBMiMo v2.5thinkingOpen×2.5—$0.019Memory (4-bit): 0.5 GB → 186 GB
Granite 4.2 8BthinkingOpen+26%—$0.024Context: 131K → 128K+1ModelScoreCostPer taskWhat changes
IK2 Horizon 7BthinkingOpen×2.3—$0.30 / $1.20Memory (4-bit): 0.5 GB → 4.5 GBIK2 Horizon 3.7BthinkingOpen+77%—$0.30 / $1.20Memory (4-bit): 0.5 GB → 2.5 GB
Gemma 4 12BthinkingOpen+61%—$0.10 / $0.30Memory (4-bit): 0.5 GB → 6 GB
Qwen 3.5 9Bnon-thinkingOpen+51%—$0.08 / $0.13Memory (4-bit): 0.5 GB → 4.8 GBQwen3.5 4BthinkingOpen+49%—$0.03 / $0.15Memory (4-bit): 0.5 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 Exam6.5#234
AA Omniscience Index-14.6#106
CritPt (physics)0#180
AA Intelligence Index 4.3.2≈8.8#235
GPQA Diamond27.8#304
Coding
Agents & Tool Use
τ²-bench · AA81#76
Writing & Language
IFBench (instructions)49.3#119
FAQ
How much does MiniCPM5 1B cost?›
There is no per-unit API price: self-host only.
Is MiniCPM5 1B free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is MiniCPM5 1B, and what is it best at?›
#232 of 334 in Chat & Knowledge, #210 of 239 in Coding, #74 of 242 in Agents & Tool Use, #118 of 245 in Writing & Language.
Is MiniCPM5 1B better than Qwen3 4B 2507 (Reasoning)?›
On AA Intelligence Index 4.3.2, MiniCPM5 1B scores 8.8 vs 8.8 — about the same.
What is the context window of MiniCPM5 1B?›
131K tokens.
When was MiniCPM5 1B released? Is there a newer version?›
Released 2026-05-25. It is the current version in its line.
Is MiniCPM5 1B open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run MiniCPM5 1B locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can MiniCPM5 1B read images?›
No, it accepts text only.
What is MiniCPM5 1B?›
MiniCPM5 1B is an open-weights language model from OpenBMB, released May 2026.
What are the best alternatives to MiniCPM5 1B?›
GPT-6 Luna (better), K2 Horizon 7B (open weights).