Ling 3.1 flash vs Qwen3.8 Flash Next
Ling 3.1 flash scores higher on the intelligence index, 41 against 40. None of the 37 machines here still sold at a published price runs Ling 3.1 flash at 32k of context. Qwen3.8 Flash Next runs on the Mac Studio M5 Ultra, 256GB, from $10,799. At 500k tokens a day that machine pays for itself in 372 years running Qwen3.8 Flash Next.
| Ling 3.1 flash | Qwen3.8 Flash Next | |
|---|---|---|
| Intelligence index | 41 | 40 |
| Class | Sonnet-class | Sonnet-class |
| Weights | 358 GB | 120 GB |
| Needs at 32k | 358 GB | 121 GB |
| Quantisation | Q4_K_M | Q4_K_M |
| Parameters | 560B (25B active) | 180B (6B active) |
| Max context | 256k | 256k |
| API price per 1M | $0.075 in / $0.225 out | $0.15 in / $0.47 out |
| Licence | MIT | qwen-community-1.0 |
| Machines here that run it | 0 of 37 | 1 of 37 |
| Cheapest machine that runs it | none listed | Mac Studio M5 Ultra, 256GB $10,799 |
Run Ling 3.1 flash in the calculator · or Qwen3.8 Flash Next on the Mac Studio M5 Ultra, 256GB
Machines that run one and not the other
Qwen3.8 Flash Next needs 121 GB of memory at 32k of context and Ling 3.1 flash needs 358 GB. That puts Qwen3.8 Flash Next on 1 of the 37 machines priced here that Ling 3.1 flash does not, starting at $10,799.
| Machine | Price | Memory | Speed on Qwen3.8 Flash Next | Pay-back |
|---|---|---|---|---|
| Mac Studio M5 Ultra, 256GB | $10,799 | 256 GB | 74 tok/s estimated | Pays back in 372 years |
Ling 3.1 flash is also head to head with GLM-5.3-Flash above it on the leaderboard. Qwen3.8 Flash Next is also head to head with DeepSeek V4-Flash below it on the leaderboard and Qwen3 235B-A22B Instruct 2507, the last-generation Qwen nearest it in size.
The assumptions behind both columns
Both columns use the same usage: 500k tokens a day at 15:1 input to output, 32k of context, $0.17 per kWh, and today's API prices held flat. Speeds marked estimated are worked out from memory bandwidth rather than measured, and pay-back scales with them. Where nobody rents an open model by the token, its API prices are the nearest hosted model's, named beside them. Machines are the 37 here with a published price that are still sold. Change any of it in the calculator.
More head to head: what Ling 3.1 flash needs · every machine that runs Qwen3.8 Flash Next · every other match-up · both against the frontier · the quickest pay-back at each level of use