Math
Like-for-like- MiniCPM5-1B
- 33.1
- Qwen3.5 397B
- 90.6
- Weighted basis
- 2 vs 2 rows
- Reading
- Qwen3.5 397B leads
MiniCPM5-1B vs Qwen3.5 397B
Updated August 22, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
Qwen3.5 397B has the higher public score, 57.49 versus 11.3, and the 90% score intervals do not overlap.
7 results are shared. Category rows based on different benchmark sets are marked directional and do not name a winner.
Which one for your work
Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.
Long documents
Prompts that approach the documented context limit
MiniCPM5-1B
MiniCPM5-1B has the larger documented context window.
Confidence: documented
Show secondary and unsupported calls
Coding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
No shared weighted benchmark basis supports a winner.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
No shared weighted benchmark basis supports a winner.
Confidence: limited
Chat turn cost
1K fresh input + 500 output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
Cache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Not enough matched evidence
The page does not recommend a cost winner because at least one model cannot fit the stated workload in one request. MiniCPM5-1B does not fit this workload in one request. Qwen3.5 397B does not fit this workload in one request. Qwen3.5 397B has no published cached-input rate, so cached tokens use its listed input rate. MiniCPM5-1B has no comparable published API token rate.
Confidence: rate-fallback
Repository review cost
50K fresh input + 3K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
7 MiniCPM5-1B only
7 Shared
31 Qwen3.5 397B only
- Shared results
- 7
- MiniCPM5-1B only
- 7
- Qwen3.5 397B only
- 31
- Like-for-like categories
- 1 / 8
2 categories use different evidence sets. Those rows remain visible for coverage context but do not name a winner.
Category results, on a stated basis
Each row states whether both averages use the same weighted benchmark set. Directional and not-comparable rows remain visible, but they never receive a winner in this template.
Knowledge
Directional only- MiniCPM5-1B
- 44.0
- Qwen3.5 397B
- 56.6
- Weighted basis
- 2 vs 4 rows
- Reading
- Directional only
Instruction following
Directional only- MiniCPM5-1B
- 58.5
- Qwen3.5 397B
- 92.6
- Weighted basis
- 2 vs 1 rows
- Reading
- Directional only
Agentic
Not comparable- MiniCPM5-1B
- Not measured
- Qwen3.5 397B
- 56.5
- Weighted basis
- 0 vs 2 rows
- Reading
- Not comparable
Coding
Not comparable- MiniCPM5-1B
- Not measured
- Qwen3.5 397B
- 66.5
- Weighted basis
- 0 vs 2 rows
- Reading
- Not comparable
Reasoning
Not comparable- MiniCPM5-1B
- Not measured
- Qwen3.5 397B
- 63.2
- Weighted basis
- 0 vs 1 rows
- Reading
- Not comparable
Multilingual
Not comparable- MiniCPM5-1B
- Not measured
- Qwen3.5 397B
- 84.7
- Weighted basis
- 0 vs 1 rows
- Reading
- Not comparable
Multimodal
Not comparable- MiniCPM5-1B
- Not measured
- Qwen3.5 397B
- 79.6
- Weighted basis
- 0 vs 2 rows
- Reading
- Not comparable
| Category | MiniCPM5-1B | Qwen3.5 397B | Weighted basis | Reading |
|---|---|---|---|---|
| Math | 33.1 | 90.6 | Like-for-like2 vs 2 rows | Qwen3.5 397B leads |
| Knowledge | 44.0 | 56.6 | Directional only2 vs 4 rows | Directional only |
| Instruction following | 58.5 | 92.6 | Directional only2 vs 1 rows | Directional only |
| Agentic | Not measured | 56.5 | Not comparable0 vs 2 rows | Not comparable |
| Coding | Not measured | 66.5 | Not comparable0 vs 2 rows | Not comparable |
| Reasoning | Not measured | 63.2 | Not comparable0 vs 1 rows | Not comparable |
| Multilingual | Not measured | 84.7 | Not comparable0 vs 1 rows | Not comparable |
| Multimodal | Not measured | 79.6 | Not comparable0 vs 2 rows | Not comparable |
Shape of the matched evidence
Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.
Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.
HMMT Feb 2026
Math
AIME26
Math
SuperGPQA
Knowledge
MMLU-Pro
Knowledge
IFEval
Instruction following
What each workload costs
Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.
Chat turn
1K fresh input + 500 output tokens
- MiniCPM5-1B
- API rate not published
- Fits in one request
- Qwen3.5 397B
- $0.0024
- Fits in one request
MiniCPM5-1B has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- MiniCPM5-1B
- API rate not published
- Fits in one request
- Qwen3.5 397B
- $0.0408
- Fits in one request
MiniCPM5-1B has no comparable published API token rate.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- MiniCPM5-1B
- API rate not published
- Does not fit in one request
- Cached-input rate unavailable
- Qwen3.5 397B
- $0.168
- Does not fit in one request
- Cached input priced at the published list-input rate
MiniCPM5-1B does not fit this workload in one request. Qwen3.5 397B does not fit this workload in one request. Qwen3.5 397B has no published cached-input rate, so cached tokens use its listed input rate. MiniCPM5-1B has no comparable published API token rate.
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationMiniCPM5-1BQwen3.5 397B
Context window
Maximum documented context; output-token limits may be lower.
MiniCPM5-1B
131K
Qwen3.5 397B
128K
API model ID
MiniCPM5-1B
Not sourced
Qwen3.5 397B
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
MiniCPM5-1B
No comparable hosted API rate
Qwen3.5 397B
Not published
Documented inputs
MiniCPM5-1B
Not sourced
Qwen3.5 397B
Not sourced
Documented outputs
MiniCPM5-1B
Not sourced
Qwen3.5 397B
Not sourced
Provider availability
MiniCPM5-1B
Not sourced
Qwen3.5 397B
Not sourced
Reasoning profile
MiniCPM5-1B
Reasoning
Qwen3.5 397B
Non-Reasoning
Weight access
MiniCPM5-1B
Open Weight
Qwen3.5 397B
Open Weight
License
MiniCPM5-1B
Open Weight
Qwen3.5 397B
Open Weight
Release date
MiniCPM5-1B
2026-05-25
Qwen3.5 397B
2026-02-16
If you already use one of these models
- Deployment change
- The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
- Quality signal
- Qwen3.5 397B has the higher public score, 57.49 versus 11.3, and the 90% score intervals do not overlap.
- Workload cost
- A complete comparable API-rate estimate is not available for both models.
- Context tradeoff
- MiniCPM5-1B has the larger documented window (131K).
Run the same representative tasks against both endpoints before changing production traffic.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence45 rows
Agentic
BFCL v4
MiniCPM5-1B25.1%SourceQwen3.5 397B—Not directly comparable
Terminal-Bench 2.0
MiniCPM5-1B—Qwen3.5 397B52.5%SourceNot directly comparable
BrowseComp
MiniCPM5-1B—Qwen3.5 397B62%SourceNot directly comparable
Claw-Eval
MiniCPM5-1B—Qwen3.5 397B56.8%SourceNot directly comparable
QwenClawBench
MiniCPM5-1B—Qwen3.5 397B51.8%SourceNot directly comparable
τ³-bench results
MiniCPM5-1B—Qwen3.5 397B68.4%SourceNot directly comparable
VITA-Bench
MiniCPM5-1B—Qwen3.5 397B43.7%SourceNot directly comparable
DeepPlanning
MiniCPM5-1B—Qwen3.5 397B37.6%SourceNot directly comparable
Toolathlon
MiniCPM5-1B—Qwen3.5 397B36.3%SourceNot directly comparable
MCP Atlas
MiniCPM5-1B—Qwen3.5 397B46.1%SourceNot directly comparable
MCP-Tasks
MiniCPM5-1B—Qwen3.5 397B74.2%SourceNot directly comparable
WideResearch
MiniCPM5-1B—Qwen3.5 397B74.0%SourceNot directly comparable
Gert Labs
MiniCPM5-1B—Qwen3.5 397B46.76%SourceNot directly comparable
ResearchClawBench
MiniCPM5-1B—Qwen3.5 397B14.2%SourceNot directly comparable
Coding
LiveCodeBench Pro
MiniCPM5-1B22.7%SourceQwen3.5 397B—Not directly comparable
LiveCodeBench v6
MiniCPM5-1B33.5%SourceQwen3.5 397B83.6%SourceQwen3.5 397B leads this result
SWE-bench Verified
MiniCPM5-1B—Qwen3.5 397B76.2%SourceNot directly comparable
SWE-bench Pro
MiniCPM5-1B—Qwen3.5 397B50.9%SourceNot directly comparable
Reasoning
Knowledge
MMLU-Pro
MiniCPM5-1B48.9%SourceQwen3.5 397B87.8%SourceQwen3.5 397B leads this result
MMLU-Redux
MiniCPM5-1B70.1%SourceQwen3.5 397B94.9%SourceQwen3.5 397B leads this result
GPQA-D
MiniCPM5-1B26.3%SourceQwen3.5 397B—Not directly comparable
SuperGPQA
MiniCPM5-1B23.1%SourceQwen3.5 397B70.4%SourceQwen3.5 397B leads this result
GPQA
MiniCPM5-1B—Qwen3.5 397B88.4%SourceNot directly comparable
C-Eval
MiniCPM5-1B—Qwen3.5 397B93%SourceNot directly comparable
HLE
MiniCPM5-1B—Qwen3.5 397B28.7%SourceNot directly comparable
Math
AIME 2025
MiniCPM5-1B40.4%SourceQwen3.5 397B—Not directly comparable
AIME26
MiniCPM5-1B40.4%SourceQwen3.5 397B93.3%SourceQwen3.5 397B leads this result
HMMT Feb 2026
MiniCPM5-1B25.8%SourceQwen3.5 397B87.9%SourceQwen3.5 397B leads this result
MATH-500
MiniCPM5-1B91.6%SourceQwen3.5 397B—Not directly comparable
HMMT Feb 2025
MiniCPM5-1B—Qwen3.5 397B94.8%SourceNot directly comparable
HMMT Nov 2025
MiniCPM5-1B—Qwen3.5 397B92.7%SourceNot directly comparable
MMAnswerBench
MiniCPM5-1B—Qwen3.5 397B80.9%SourceNot directly comparable
Multilingual
Multimodal
MMMU-Pro
MiniCPM5-1B—Qwen3.5 397B79%SourceNot directly comparable
MathVision
MiniCPM5-1B—Qwen3.5 397B88.6%SourceNot directly comparable
CharXiv
MiniCPM5-1B—Qwen3.5 397B80.8%SourceNot directly comparable
VideoMMMU
MiniCPM5-1B—Qwen3.5 397B84.7%SourceNot directly comparable
ScreenSpot Pro
MiniCPM5-1B—Qwen3.5 397B65.6%SourceNot directly comparable
V*
MiniCPM5-1B—Qwen3.5 397B95.8%SourceNot directly comparable
Instruction following
Frequently asked questions
Which is better, MiniCPM5-1B or Qwen3.5 397B?
Qwen3.5 397B has the higher public score, 57.49 versus 11.3, and the 90% score intervals do not overlap. That is the clearest overall quality signal in the current public data.
Which is better for coding, MiniCPM5-1B or Qwen3.5 397B?
The published evidence does not provide a shared weighted coding basis for both models, so BenchLM does not name a coding winner.
Which is better for agentic tasks, MiniCPM5-1B or Qwen3.5 397B?
The published evidence does not provide a shared weighted agentic tasks basis for both models, so BenchLM does not name a agentic tasks winner.
Which costs less, MiniCPM5-1B or Qwen3.5 397B?
Both models do not have comparable published API token rates, so this page does not name a universal price winner.
Which has the larger context window, MiniCPM5-1B or Qwen3.5 397B?
MiniCPM5-1B has the larger documented context window: 131K, compared with 128K.
Related comparisons
Last updated August 22, 2026
Watch MiniCPM5-1B vs Qwen3.5 397B
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.