ModelsCompareBest forBenchmarksStatusPricingAPI
Compare / head-to-head

ERNIE 5.0vsStep 3.5 Flash

Step 3.5 Flash leads 4 of 4 shared benchmarks. Step 3.5 Flash has the larger context window (256K tokens).

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

Shared benchmarks
0 – 4
Step 3.5 Flash leads
Cheaper per token
list price, input + output
Larger context
Step 3.5 Flash
256K tokens
Provider uptime (30d)
— · —
Baidu · StepFun
ERNIE 5.0
Baidu · released 2026-02-06
textvision
Context
128K
Max out
66K
Input /1M
Output /1M
Cached /1M
Scores
21 · 5 core
Step 3.5 Flash
StepFun · released 2026-02-12
text
Context
256K
Max out
Input /1M
Output /1M
Cached /1M
Scores
18 · 5 core
Quality

Benchmark matrix

BenchmarkERNIE 5.0Step 3.5 FlashΔEdge
Reported by both · 4
AIME 202589.06% 97.3% -8.2 ptStep 3.5 Flash
BrowseComp-ZH64.71% 66.9% -2.2 ptStep 3.5 Flash
LiveCodeBench v676.21% 86.4% -10.2 ptStep 3.5 Flash
Tau2-Bench78.79% 88.2% -9.4 ptStep 3.5 Flash
Only ERNIE 5.0 reports · 17
GPQA Diamond86.36% not reported
MMLU-Pro83.8% not reported
ACEBench Chinese89.6% not reported
ACEBench English87.7% not reported
BBEH66.63% not reported
BFCL v466.47% not reported
ChineseSimpleQA86.03% not reported
HMMT 202579.58% not reported
HumanEval+94.48% not reported
Humanity's Last Exam25.81% not reported
IFEval93.35% not reported
MBPP+82.54% not reported
Multi-IF85.56% not reported
MultiChallenge65.98% not reported
SimpleQA74.01% not reported
SpreadsheetBench40.08% not reported
ZebraLogic96.5% not reported
Only Step 3.5 Flash reports · 13
SWE-bench Verifiednot reported74.4%
AIME 2026not reported96.7% matharena
LMArena Elonot reported1403.7
BrowseCompnot reported51.6%
BrowseComp (with context manager)not reported69%
BrowseComp-ZH (with context manager)not reported73.7%
GAIA (no file)not reported84.5%
HMMT 2025 (Feb/Nov average)not reported96.2%
IMOAnswerBenchnot reported85.4%
ResearchRubricsnot reported65.3%
Terminal-Bench 2.0not reported51%
xbench-DeepSearch 2025.05not reported83.7%
xbench-DeepSearch 2025.10not reported56.3%
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 Step 3.5 Flash 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 5.0Step 3.5 FlashEdge
Context window128K tokens256K tokensStep 3.5 Flash
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
Released2026-02-062026-02-12
Cited benchmark scores2118
Reliability

Provider status

All providers
More matchups

Step 3.5 Flash vs …

Built by Respan
Which one wins on your data?

Public benchmarks are a starting point. Run ERNIE 5.0 and Step 3.5 Flash on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.