Is Step 3.5 Flash a good deal?
Whether anything in this catalogue beats Step 3.5 Flash on both quality and price, and what you give up if it does. A computation on the current catalogue, not an opinion.
12 undominated of 145 · Oct 6, 2026
As of Oct 6, 2026, Step 3.5 Flash is dominated for Balanced on LMArena. Gemma 4 26B A4B scores 30.9 higher and costs 16% less, with a covering envelope. 12 of 145 rated, priced standard models are undominated.
Inspect model evidence Compare differences & requirements
Gemma 4 26B A4B is both better and cheaper than Step 3.5 Flash: 30.9 points higher on LMArena and 16% less per million tokens, $0.023 cheaper at this mix.
LMArena Elo · higher is better
Scale starts at 1040 Elo
1403.0
Effective $/M · Balanced · lower is better
$0.150/M
LMArena Elo · higher is better
Scale starts at 1040 Elo
1433.9
Effective $/M · Balanced · lower is better
$0.127/M
Step 3.5 Flash takes text, returns up to 65,536 tokens from a 262,144-token context, and is listed by 1 seller.
Compared against 145 rated, priced models on this lens: 2 models dominate it and give up nothing, none does so with a trade-off. Step 3.5 Flash scores 1403 at $0.150 per million tokens for this mix.
Envelope-safe replacements
Each row scores at least as high, costs no more, and covers this model’s context, output, modalities, tools, and reasoning. A cheaper narrower model is not listed here.
| Model | LMArena | Effective $/M | You save |
|---|---|---|---|
| Gemma 4 26B A4B | 1433.9 +30.9 | $0.127/M | 16% |
| DeepSeek V4 Flash 0423 | 1432.1 +29.1 | $0.113/M | 25% |
The link keeps the model and mix, using the current catalogue. Monthly spend and switching cost stay in this browser tab and are left out of the link.
Saved decision references
Save the model, workload, catalogue date and capability-preservation rule in this browser. Spend, switching cost and bill contents are not saved. No account or notifications.
Constraint: replacements must preserve the model’s capabilities; any losses remain named trade-offs.
Historical decisions cannot be fully replayed from saved references: past prices, scores and capability evidence are not stored.