Compare · ModelsLive · 2 picked · head to head
gpt-oss-120b (free) vs Step 3.5 Flash
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Step 3.5 Flash wins on 9/9 benchmarks
Step 3.5 Flash wins 9 of 9 shared benchmarks. Leads in speed · math · knowledge.
Category leads
speed·Step 3.5 Flashmath·Step 3.5 Flashknowledge·Step 3.5 Flashlanguage·Step 3.5 Flashcoding·Step 3.5 Flash
Hype vs Reality
Attention vs performance
gpt-oss-120b (free)
#22 by perf·no signal
Step 3.5 Flash
#9 by perf·#11 by attention
Best value
Step 3.5 Flash
gpt-oss-120b (free)
—
$0.00/M
Step 3.5 Flash
384.5 pts/$
$0.20/M
Vendor risk
Mixed exposure
One or more vendors flagged
OpenAI
$840.0B·Tier 1
StepFun
$5.0B·Tier 1
Head to head
9 benchmarks · 2 models
gpt-oss-120b (free)Step 3.5 Flash
Artificial Analysis · Agentic Index
Step 3.5 Flash leads by +14.1
Artificial Analysis Agentic Index · a composite score measuring how well a model performs in agentic workflows · multi-step tool use, planning, error recovery, and autonomous task completion. Aggregates results from multiple agentic benchmarks including SWE-bench, tool-use tests, and planning evaluations. The canonical single-number metric for "how good is this model as an agent?"
gpt-oss-120b (free)
37.9
Step 3.5 Flash
52.0
Artificial Analysis · Coding Index
Step 3.5 Flash leads by +3.0
Artificial Analysis Coding Index · a composite score that aggregates performance across multiple coding benchmarks into a single index. Tracks code generation quality, debugging ability, multi-language competence, and real-world software engineering tasks. Used by Artificial Analysis to rank model coding capability in a normalized, comparable format. Useful for developers choosing between models for coding-heavy workloads.
gpt-oss-120b (free)
28.6
Step 3.5 Flash
31.6
Artificial Analysis · Quality Index
Step 3.5 Flash leads by +4.5
gpt-oss-120b (free)
33.3
Step 3.5 Flash
37.8
OpenCompass · AIME2025
Step 3.5 Flash leads by +2.3
gpt-oss-120b (free)
93.4
Step 3.5 Flash
95.7
OpenCompass · GPQA-Diamond
Step 3.5 Flash leads by +4.8
gpt-oss-120b (free)
78.9
Step 3.5 Flash
83.7
OpenCompass · HLE
Step 3.5 Flash leads by +3.3
gpt-oss-120b (free)
18.3
Step 3.5 Flash
21.6
OpenCompass · IFEval
Step 3.5 Flash leads by +3.0
gpt-oss-120b (free)
90.2
Step 3.5 Flash
93.2
OpenCompass · LiveCodeBenchV6
Step 3.5 Flash leads by +5.5
gpt-oss-120b (free)
78.4
Step 3.5 Flash
83.9
OpenCompass · MMLU-Pro
Step 3.5 Flash leads by +3.8
gpt-oss-120b (free)
79.7
Step 3.5 Flash
83.5
Full benchmark table
| Benchmark | gpt-oss-120b (free) | Step 3.5 Flash |
|---|---|---|
Artificial Analysis · Agentic Index Artificial Analysis Agentic Index · a composite score measuring how well a model performs in agentic workflows · multi-step tool use, planning, error recovery, and autonomous task completion. Aggregates results from multiple agentic benchmarks including SWE-bench, tool-use tests, and planning evaluations. The canonical single-number metric for "how good is this model as an agent?" | 37.9 | 52.0 |
Artificial Analysis · Coding Index Artificial Analysis Coding Index · a composite score that aggregates performance across multiple coding benchmarks into a single index. Tracks code generation quality, debugging ability, multi-language competence, and real-world software engineering tasks. Used by Artificial Analysis to rank model coding capability in a normalized, comparable format. Useful for developers choosing between models for coding-heavy workloads. | 28.6 | 31.6 |
Artificial Analysis · Quality Index | 33.3 | 37.8 |
OpenCompass · AIME2025 | 93.4 | 95.7 |
OpenCompass · GPQA-Diamond | 78.9 | 83.7 |
OpenCompass · HLE | 18.3 | 21.6 |
OpenCompass · IFEval | 90.2 | 93.2 |
OpenCompass · LiveCodeBenchV6 | 78.4 | 83.9 |
OpenCompass · MMLU-Pro | 79.7 | 83.5 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.00 | $0.00 | 131K tokens (~66 books) | — | |
| $0.10 | $0.30 | 262K tokens (~131 books) | $1.50 |