JT-35B-Flash vs MiniCPM5-1B: Benchmarks, Pricing & Capabilities Compared
TL;DR — JT-35B-Flash wins for reasoning + long-context · MiniCPM5-1B wins for general use.
JT-35B-Flash China Mobile
- Released
- 2026-05-14
- Context window
- 256K tokens
- Input price
- $0.00 / Mtok
- Output price
- $0.00 / Mtok
MiniCPM5-1B OpenBMB
- Released
- 2026-05-25
- Context window
- 128K tokens
- Input price
- $0.00 / Mtok
- Output price
- $0.00 / Mtok
Benchmark comparison
| Benchmark | JT-35B-Flash | MiniCPM5-1B |
|---|---|---|
| AA Intelligence Index | 36.1 ✓ | 17.9 |
| GPQA Diamond | 82.9% ✓ | 26.9% |
| HLE | 6.1% ✓ | 4.6% |
| IF-Bench | 42.0% ✓ | 35.2% |
| LiveCodeBench Reasoning | 55.3% ✓ | 4.7% |
| SciCode | 29.1% ✓ | 1.4% |
| TAU2-bench | 99.1% ✓ | 82.5% |
| TerminalBench-Hard | 28.8% ✓ | 0.0% |
Pricing comparison
| Metric | JT-35B-Flash | MiniCPM5-1B |
|---|---|---|
| Input ($/Mtok) | $0.00 | $0.00 |
| Output ($/Mtok) | $0.00 | $0.00 |
| Cached input ($/Mtok) | — | — |
| Cost per 1M-token roundtrip (1M in + 1M out) | $0.00 | $0.00 |
Context window & modalities
| Attribute | JT-35B-Flash | MiniCPM5-1B |
|---|---|---|
| Context window | 256K tokens | 128K tokens |
| Input modalities | text | text |
| Output modalities | text | text |
| Knowledge cutoff | — | — |
Verdict by use case
Coding
Insufficient data
Basis: SWE-bench
No shared coding benchmark.
Reasoning
→ JT-35B-Flash
Basis: GPQA Diamond
JT-35B-Flash 82.9% vs MiniCPM5-1B 26.9% on GPQA Diamond.
Math
Insufficient data
Basis: MATH / AIME
No shared math benchmark.
Long context
→ JT-35B-Flash
Basis: Context window
JT-35B-Flash 256K tokens vs MiniCPM5-1B 128K tokens.
Cost
Tie
Basis: Input $/Mtok
JT-35B-Flash $0/Mtok vs MiniCPM5-1B $0/Mtok input.
Changelog & releases
JT-35B-Flash
Released 2026-05-14
MiniCPM5-1B
Released 2026-05-25