Inkling 2 Small vs Qwen3 VL 30B A3B Thinking
Prices checked
Qwen3 VL 30B A3B Thinking takes 3 of 5 rounds against Inkling 2 Small. It is also the cheaper of the two to run, at $0.18 per million tokens against $0.24.
Thinking MachinesInkling 2 Small
1M context · tool calling · reads images · open weights
$0.24/M tokensCheapest price, at Kilo Gateway · $0.12 per 1,000 pages
Alibaba CloudWins 3 roundsQwen3 VL 30B A3B Thinking
262k context · tool calling · reads images · open weights
$0.18/M tokensCheapest price, at SiliconFlow (China) · 8.8¢ per 1,000 pages
Maker's price in (lower wins)
$0.30/M
$0.29/M
Maker's price out (lower wins)
$1/M
$1/M
Cheapest price (lower wins)
$0.24/M at Kilo Gateway
$0.18/M at SiliconFlow (China)
Maker's price, blended (lower wins)
$0.47/M
$0.47/M
Cost per 1,000 pages (lower wins)
$0.12
8.8¢
Context
1M tokens
262k tokens
Open weights
yes
yes
Tool calling
yes
yes
Structured output
yes
yes
Reads images
yes
yes
Reasoning
yes
yes
Released
9 Nov 2026
6 Oct 2025
Providers selling it
14
8
Price we measure against
OpenRouter
OpenRouter
Trains on your prompts (no is better)
not published
not published
Providers selling both · 3
Kilo Gateway$0.24 / $0.47
Nous Portal$0.47 / $0.47
OpenRouter$0.47 / $0.47
Inkling 2 Small, then Qwen3 VL 30B A3B Thinking, per million tokens.
Go further
Add a third or fourth model, or change the weighting, in the tool.
Frequently asked
- Which is cheaper, Inkling 2 Small or Qwen3 VL 30B A3B Thinking?
- Qwen3 VL 30B A3B Thinking, at $0.18 per million tokens against $0.24 for Inkling 2 Small, blended 3 in : 1 out, read on 13 Nov 2026.
- Where is each one cheapest?
- Inkling 2 Small: $0.24 at Kilo Gateway. Qwen3 VL 30B A3B Thinking: $0.18 at SiliconFlow (China).
- Can I buy both from one provider?
- Yes. 3 providers sell both: Kilo Gateway, Nous Portal, OpenRouter. The cheapest of them for the two together is Kilo Gateway.
- Which holds more context?
- Inkling 2 Small, at 1M tokens against 262k.
Prices exclude tax, read 13 Nov 2026 from each provider's own catalog. They are per M tokens, blended 3 in : 1 out, and a page is 600 tokens in and 60 out. Bars are scaled to the larger value in each row.
Related comparisons