LLMs TigerLive LLM benchmarks

Step 3.7 Flash: benchmark scores, price and ranking

Step 3.7 Flash from StepFun ranks #469 of 743 LLMs on the LLMs Tiger aggregate benchmark score (47.0), based on 3 benchmark results captured up to 2026-10-13.

Organisation
StepFun
Type
Open-weight / local model
Open weights
Yes
Released
2026-05-29 (source: epoch)
Cheapest API price
$0.200 in / $1.15 out per million tokens (via deepinfra, observed 2026-10-13T20:47Z)
Aggregate rank
#469 of 743 (score 47.0)

Relative to other models, Step 3.7 Flash is strongest on CritPt (#60 of 115) and weakest on SciCode (#67 of 106).

Across the 3 benchmarks where at least ten models are scored, Step 3.7 Flash is in the top 10% on 0 and in the top half on 0.

Its cheapest listed input price, $0.200 per million tokens, is lower than that of 67% of the 427 models with a listed price.

Step 3.7 Flash benchmark results

Step 3.7 Flash benchmark scores, ranked against all models measured on each benchmark
BenchmarkScoreRankSourceCaptured
CritPt2.3%#60 of 115Epoch AI Benchmarking Hub2026-10-13
ALE-Bench694#44 of 79Epoch AI Benchmarking Hub2026-10-13
SciCode40.0%#67 of 106Epoch AI Benchmarking Hub2026-10-13

About Step 3.7 Flash

Step 3.7 Flash is an open-weight model that can run on your own hardware from StepFun. It was released on 2026-05-29, according to Epoch AI. The cheapest API listing we track charges $0.200 per million input tokens and $1.15 per million output tokens, through deepinfra. We hold 3 benchmark results for it across 2 areas, and none is in the top tenth of its board.

Where Step 3.7 Flash stands

On our aggregate score, Step 3.7 Flash is #469 of 743 ranked models, in the lower half. Its score is 47.0; the scale is centred on 50, so it sits 3.0 points below the typical model. That beats 37% of the ranked field. Among the 2 ranked StepFun models it is #2. The next one up is step-5-preview-high at 50.0.

Step 3.7 Flash, benchmark by benchmark

Step 3.7 Flash lands below the middle on CritPt: 2.3%, #60 of 115, with 51% of the field scoring higher. GPT-5.6 Sol (max) leads at 32.3%; the closest result above is Muse Glimmer (high) at 2.6%. It covers unpublished research-level physics problems. The board median is 2.9%, so Step 3.7 Flash is 0.6 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

Step 3.7 Flash lands below the middle on ALE-Bench: 694, #44 of 79, with 54% of the field scoring higher. GPT-6 Astra (max) leads at 2,951; the closest result above is GPT-5 nano (high) at 719. It covers AtCoder heuristic-optimisation contests (performance rating). The board median is 764, so Step 3.7 Flash is 70 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

Step 3.7 Flash lands below the middle on SciCode: 40.0%, #67 of 106, with 62% of the field scoring higher. Claude Opus 5.5 (max) leads at 66.9%; the closest result above is Gemma 4 26B A4B at 40.0%. It covers scientific research code. The board median is 44.3%, so Step 3.7 Flash is 4.3 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

Strengths and weaknesses

Software engineering: uneven, ahead of 41% of the field on average across 2 benchmarks. Its best board here is ALE-Bench (#44 of 79) and its weakest is SciCode (#67 of 106).

Price and value

The median input price across the 427 priced models is $0.500 per million tokens, so Step 3.7 Flash costs less than the typical listing. Models that score within 1.5 points or better and cost less: Qwen2.5-Coder-7B-Instruct ($0.010, 49.5), DeepSeek V4.1 Flash (high) ($0.010, 48.9), Qwen2.5-Coder-3B-Instruct ($0.010, 46.5). Similar scores cost more on MiniMax-M3 ($0.230) and DeepSeek-V3.1 ($0.250).

Reading these numbers

The 3 scores on this page come from 1 source: Epoch AI Benchmarking Hub (3). No official leaderboard supplied any of them, so treat them as third-party measurements. With 3 results the aggregate rank can move a lot when one new score arrives.

Models ranked nearby

More from StepFun

Compare it on the live leaderboard → · How the aggregate score works