Skip to main content

Model profile

MiniCPM5-1B

OpenBMBCurrentReleased May 25, 2026
Data verified
Overall Score
Unranked
Arena Elo
Not listed
Eligible category ranks
2of 8
Price (1M tokens)
Not listedAPI pricing
Speed
Not listed
Context
131K

Evidence coverage

14 of 323 tracked benchmarks are published. 14 are verified and 0 provisional. 6 of 8 categories are measured.

Updated July 23, 2026Methodology
Published / tracked
14 / 323
Verified
14
Provisional
0
Categories with evidence
6 / 8

Evidence by category

  • Agentic1 benchmark
    Verified
  • Coding2 benchmarks
    Verified
  • Reasoning1 benchmark
    Verified
  • Knowledge4 benchmarks
    Verified
  • Math4 benchmarks
    Verified
  • Multilingual0 benchmarks
    Not measured
  • Multimodal0 benchmarks
    Not measured
  • Inst. Following2 benchmarks
    Verified
Open WeightSelf-hostReasoning
Confidence:
Medium
1b

BenchLM is tracking MiniCPM5-1B, but this profile is currently excluded from the public leaderboard because it still lacks enough non-generated benchmark coverage to rank safely. Only non-generated public benchmark rows appear below.

MiniCPM5-1B is a open weight model with a 131K token context window. It uses explicit chain-of-thought reasoning, which typically improves performance on math and complex reasoning tasks at the cost of higher latency and token usage.

This profile currently has 14 of 323 tracked benchmarks. BenchLM only exposes non-generated benchmark rows publicly, so missing categories stay blank until a sourced evaluation is available.

Its strongest category is Instruction Following (#30), while its weakest is Agentic (#71). This performance profile makes it a well-rounded choice across a range of tasks.

Peer position

Exact provisional scores and ranks for the closest listed peers. A score can appear before a model clears the evidence threshold for a rank, so equal scores can have different rank states.

Range 77.4483.93

  1. Claude Mythos 5
    Anthropic
    #183.93
    Claude Mythos 5 is #1 with a score of 83.93.
    Compare
  2. Claude Fable 5
    Anthropic
    #283.68
    Claude Fable 5 is #2 with a score of 83.68.
    Compare
  3. GPT-5.6 Sol
    OpenAI
    #381.96
    GPT-5.6 Sol is #3 with a score of 81.96.
    Compare
  4. Kimi K3
    Moonshot AI
    #480.96
    Kimi K3 is #4 with a score of 80.96.
    Compare
  5. Claude Opus 4.8
    Anthropic
    #578.34
    Claude Opus 4.8 is #5 with a score of 78.34.
    Compare
  6. Muse Spark 1.1
    Meta
    #677.44
    Muse Spark 1.1 is #6 with a score of 77.44.
    Compare
  7. MiniCPM5-1BCurrent model
    OpenBMB
    UnrankedNot measured
    MiniCPM5-1B is Unranked with a score of Not measured.

Category percentile

More

Relative position among models eligible for each sourced category. A higher percentile means a stronger position within that category's ranked cohort; 100 is highest.

  1. Inst. Following3%
    Eligible cohort rank #30 of 31Category score 24.7
  2. Agentic41%
    Eligible cohort rank #71 of 119Category score 45.7

Category evidence

Scores and ranks appear only where this model has published benchmark evidence. Categories without displayable source records remain not measured.

Category scores, ranks, weighting, benchmark coverage, and evidence status
CategoryScore
AgenticRank #71 of 119Percentile 41stWeight 22%1 benchmarkVerified45.7
CodingWeight 20%2 benchmarksVerifiedScore pending
ReasoningWeight 17%1 benchmarkVerifiedScore pending
KnowledgeRank Not rankedWeight 12%4 benchmarksVerified39.5
MathRank Not rankedWeight 5%4 benchmarksVerified5.3
MultilingualWeight 7%0 benchmarksNot measuredNot measured
MultimodalWeight 12%0 benchmarksNot measuredNot measured
Inst. FollowingRank #30 of 31Percentile 3rdWeight 5%2 benchmarksVerified24.7

Benchmark Details

Rows below have a displayable published verification record. Each source link and provenance note remains in the page HTML while its category is closed. Source-unverified manual rows and generated rows stay hidden.

Agentic1 benchmark
BFCL v4Provider exact

Berkeley Function Calling Leaderboard v4

25.1%Display only
Source: OpenBMB MiniCPM5-1B model cardProvenance: OpenBMB reports MiniCPM5-1B thinking mode at 25.15 on BFCLv4 in its public evaluation table.
Coding2 benchmarks
LiveCodeBench ProProvider exact
22.7%Display only
Source: OpenBMB MiniCPM5-1B model cardProvenance: OpenBMB reports MiniCPM5-1B thinking mode at 22.68 on LCB-Pro 25Q2 Easy. BenchLM stores this on the display-only LiveCodeBench Pro key.
LiveCodeBench v6Provider exact
33.5%Display only
Source: OpenBMB MiniCPM5-1B model cardProvenance: OpenBMB reports MiniCPM5-1B thinking mode at 33.52 on LCB-v6 @Avg3. BenchLM stores this on the display-only LiveCodeBench v6 key.
Reasoning1 benchmark
BBHProvider exact

BIG-Bench Hard

71.9%Display only
Source: OpenBMB MiniCPM5-1B model cardProvenance: OpenBMB reports MiniCPM5-1B thinking mode at 71.89 on BBH in its public evaluation table.
Knowledge4 benchmarks
MMLU-ProProvider exact

Massive Multitask Language Understanding Professional

48.9%Weighted 30%
Source: OpenBMB MiniCPM5-1B model cardProvenance: OpenBMB reports MiniCPM5-1B thinking mode at 48.85 on MMLU-Pro in its public evaluation table.
SuperGPQAProvider exact

SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines

23.1%Weighted 7%
Source: OpenBMB MiniCPM5-1B model cardProvenance: OpenBMB reports MiniCPM5-1B thinking mode at 23.14 on SuperGPQA in its public evaluation table.
MMLU-ReduxProvider exact
70.1%Display only
Source: OpenBMB MiniCPM5-1B model cardProvenance: OpenBMB reports MiniCPM5-1B thinking mode at 70.06 on MMLU-Redux in its public evaluation table.
GPQA-DProvider exact

GPQA Diamond

26.3%Display only
Source: OpenBMB MiniCPM5-1B model cardProvenance: OpenBMB reports MiniCPM5-1B thinking mode at 26.26 on GPQA-Diamond in its public evaluation table.
Math4 benchmarks
AIME26Provider exact

AIME 2026

40.4%Weighted 25%
Source: OpenBMB MiniCPM5-1B model cardProvenance: OpenBMB reports MiniCPM5-1B thinking mode at 40.42 on AIME-2026 @Avg16 in its public evaluation table.
HMMT Feb 2026Provider exact

Harvard-MIT Mathematics Tournament February 2026

25.8%Weighted 25%
Source: OpenBMB MiniCPM5-1B model cardProvenance: OpenBMB reports MiniCPM5-1B thinking mode at 25.76 on HMMT Feb 2026 @Avg16 in its public evaluation table.
AIME 2025Provider exact

American Invitational Mathematics Examination 2025

40.4%Display only
Source: OpenBMB MiniCPM5-1B model cardProvenance: OpenBMB reports MiniCPM5-1B thinking mode at 40.42 on AIME-2025 @Avg16 in its public evaluation table.
MATH-500Provider exact

MATH-500 Problem Set

91.6%Display only
Source: OpenBMB MiniCPM5-1B model cardProvenance: OpenBMB reports MiniCPM5-1B thinking mode at 91.60 on MATH-500 in its public evaluation table.
Inst. Following2 benchmarks
IFBenchProvider exact

Instruction Following Benchmark

46.7%Weighted 65%
Source: OpenBMB MiniCPM5-1B model cardProvenance: OpenBMB reports MiniCPM5-1B thinking mode at 46.67 on IFBench in its public evaluation table.
IFEvalProvider exact

Instruction-Following Eval

80.4%Weighted 35%
Source: OpenBMB MiniCPM5-1B model cardProvenance: OpenBMB reports MiniCPM5-1B thinking mode at 80.41 on IFEval in its public evaluation table.

Frequently Asked Questions

How does MiniCPM5-1B perform overall in AI benchmarks?

MiniCPM5-1B has 14 published benchmark scores on BenchLM, but it does not yet have enough non-generated coverage to receive a global overall rank.

Is MiniCPM5-1B good for knowledge and understanding?

MiniCPM5-1B has visible benchmark coverage in knowledge and understanding, but BenchLM does not currently assign it a global category rank there.

Is MiniCPM5-1B good for coding and programming?

MiniCPM5-1B has visible benchmark coverage in coding and programming, but BenchLM does not currently assign it a global category rank there.

Is MiniCPM5-1B good for mathematics?

MiniCPM5-1B has visible benchmark coverage in mathematics, but BenchLM does not currently assign it a global category rank there.

Is MiniCPM5-1B good for reasoning and logic?

MiniCPM5-1B has visible benchmark coverage in reasoning and logic, but BenchLM does not currently assign it a global category rank there.

Is MiniCPM5-1B good for agentic tool use and computer tasks?

MiniCPM5-1B ranks #71 out of 119 models in agentic tool use and computer tasks benchmarks with an average score of 45.7. There are stronger options in this category.

Is MiniCPM5-1B good for instruction following?

MiniCPM5-1B ranks #30 out of 31 models in instruction following benchmarks with an average score of 24.7. There are stronger options in this category.

Is MiniCPM5-1B open source?

Yes, MiniCPM5-1B is an open weight model created by OpenBMB, meaning it can be downloaded and run locally or fine-tuned for specific use cases.

Does MiniCPM5-1B have full benchmark coverage on BenchLM?

Not yet. MiniCPM5-1B currently has 14 published benchmark scores out of the 323 benchmarks BenchLM tracks. BenchLM only exposes non-generated public benchmark rows, so missing categories stay blank until a sourced evaluation is available.

What is the context window size of MiniCPM5-1B?

MiniCPM5-1B has a published context window of 131K, which determines how much text it can process in a single interaction.

Last updated: July 23, 2026 · Runtime metrics stay blank until BenchLM has a sourced snapshot.

Choose with this week’s evidence

Join 2,000+ readers for ranking moves, new releases, pricing changes, and the evidence behind them.

Free. One email per week.