ModelsCompareBest forBenchmarksStatusPricingAPI
Compare / head-to-head

ERNIE 5.0vsLongCat 2.0

ERNIE 5.0 and LongCat 2.0 split their 2 shared benchmarks evenly. LongCat 2.0 has the larger context window (1M tokens).

Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.

Shared benchmarks
1 – 1
even split
Cheaper per token
list price, input + output
Larger context
LongCat 2.0
1M tokens
Provider uptime (30d)
— · —
Baidu · Meituan LongCat
ERNIE 5.0
Baidu · released 2026-02-06
textvision
Context
128K
Max out
66K
Input /1M
Output /1M
Cached /1M
Scores
21 · 5 core
LongCat 2.0
Meituan LongCat · released 2026-06-30
text
Context
1M
Max out
Input /1M
Output /1M
Cached /1M
Scores
10 · 4 core
Quality

Benchmark matrix

BenchmarkERNIE 5.0LongCat 2.0ΔEdge
Reported by both · 2
GPQA Diamond86.36% 88.9% -2.5 ptLongCat 2.0
IFEval93.35% 90% +3.3 ptERNIE 5.0
Only ERNIE 5.0 reports · 19
MMLU-Pro83.8% not reported
AIME 202589.06% not reported
ACEBench Chinese89.6% not reported
ACEBench English87.7% not reported
BBEH66.63% not reported
BFCL v466.47% not reported
BrowseComp-ZH64.71% not reported
ChineseSimpleQA86.03% not reported
HMMT 202579.58% not reported
HumanEval+94.48% not reported
Humanity's Last Exam25.81% not reported
LiveCodeBench v676.21% not reported
MBPP+82.54% not reported
Multi-IF85.56% not reported
MultiChallenge65.98% not reported
SimpleQA74.01% not reported
SpreadsheetBench40.08% not reported
Tau2-Bench78.79% not reported
ZebraLogic96.5% not reported
Only LongCat 2.0 reports · 8
BrowseCompnot reported79.9%
FORTEnot reported73.2%
IMO-AnswerBenchnot reported81.8%
RWSearchnot reported78.8%
SWE-bench Multilingualnot reported77.3%
SWE-bench Pronot reported59.5%
Terminal-Bench 2.1not reported70.8%
Writing Benchnot reported83.8%
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is ERNIE 5.0 minus LongCat 2.0 in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing

Side by side

SpecERNIE 5.0LongCat 2.0Edge
Context window128K tokens1M tokensLongCat 2.0
Max output66K tokens
Input price / 1M
Output price / 1M
Cached input / 1M
Input + output / 1M
Lower is cheaper. List prices; batch, tool and regional fees excluded.
Modalitiestext · visiontextERNIE 5.0
Released2026-02-062026-06-30
Cited benchmark scores2110
Reliability

Provider status

All providers
Built by Respan
Which one wins on your data?

Public benchmarks are a starting point. Run ERNIE 5.0 and LongCat 2.0 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.