Qwen3.8 27B vs Qwen3.6 27B API pricing
Updated
Qwen3.8 27B: USD 0.039 input, USD 1.25 output per 1M tokens (Chutes). Qwen3.6 27B: USD 0.25 input, USD 2 output per 1M tokens (Chutes). At 3 input tokens per output token, Qwen3.6 27B costs 2.01× as much as Qwen3.8 27B (USD 0.6875 vs USD 0.3417 per 1M).
How do Qwen3.8 27B and Qwen3.6 27B differ on price?
- Cheapest input price: Qwen3.8 27B USD 0.039 against USD 0.25 per 1M tokens, so Qwen3.6 27B costs 6.4× as much.
- Cheapest output price: Qwen3.8 27B USD 1.25 against USD 1.95 per 1M tokens, so Qwen3.6 27B costs 1.56× as much (56% more).
- Among the 201 indexed models, Qwen3.8 27B ranks 74th cheapest and Qwen3.6 27B 108th on the 3:1 blend.
- No input:output mix changes the answer: Qwen3.8 27B is cheaper at every mix before cache discounts.
- Both list cached-input discounts: Qwen3.8 27B at 14 standard prices of 19, from USD 0.0039 (Chutes, 90% below that provider's input price); Qwen3.6 27B at 3 standard prices of 8, from USD 0.025 (Chutes, 90% below that provider's input price).
- Qwen3.8 27B states up to 1,000,000 tokens of context, 3.8× the 262,144 of Qwen3.6 27B.
- Qwen3.8 27B has 19 providers with a standard price, 11 more than Qwen3.6 27B; 7 providers sell both (Chutes, Reka, DeepInfra, Featherless and 3 more).
- On one account, Chutes charges USD 0.3417 for Qwen3.8 27B (its cheapest price) and USD 0.6875 for Qwen3.6 27B (its cheapest price) at 3:1, the lowest combined price among the providers that sell both.
- Shopping around matters more for Qwen3.8 27B: its dearest standard price is 4.7× its cheapest, against 1.96× for Qwen3.6 27B.
- Qwen's own API charges USD 1.125 for Qwen3.8 27B at 3:1, 229% above the cheapest provider (Chutes).
- Qwen's own API charges USD 1.35 for Qwen3.6 27B at 3:1, 96% above the cheapest provider (Chutes).
- Qwen3.8 27B has a batch price at 1 provider, from USD 0.5625 at 3:1 (Doubleword); Qwen3.6 27B has no batch price on file.
- For the one-developer coding agent preset, the gap is USD 1,881 a year in Qwen3.8 27B's favour (Qwen3.8 27B USD 129.00 a month, Qwen3.6 27B USD 285.75).
- Per character of input, Qwen3.8 27B costs 84% less than Qwen3.6 27B in every measured language, because both use the Qwen3.8 27B tokenizer.
Qwen3.8 27B and Qwen3.6 27B prices side by side
USD per 1M tokens, standard tier. "Cheapest" is the lowest price at 3 input tokens to 1 output token; each line can come from a different provider.
| Price | Qwen3.8 27B | Qwen3.6 27B |
|---|---|---|
| Model page | Qwen3.8 27B Qwen · 19 providers | Qwen3.6 27B Qwen · 8 providers |
| Cheapest standard price, blended 3:1 | USD 0.3417 lower USD 0.039 in / USD 1.25 out at Chutes Chutes page checked 2026-11-15 | USD 0.6875 USD 0.25 in / USD 2 out at Chutes Chutes page checked 2026-11-15 |
| Cheapest input | USD 0.039 lower Chutes source | USD 0.25 Chutes source |
| Cheapest cached input | USD 0.0039 lower Chutes source | USD 0.025 Chutes source |
| Cheapest output | USD 1.25 lower Chutes source | USD 1.95 Reka source |
| Maker's own API | USD 0.50 / USD 3 Alibaba Cloud source | USD 0.60 / USD 3.60 Alibaba Cloud source |
| Median standard price | USD 0.40 / USD 2.50 input / output, 19 prices | USD 0.32 / USD 3.20 input / output, 8 prices |
| Context window | 1,000,000 tokens, largest stated; from 131,072 at some providers | 262,144 tokens, largest stated; from 32,768 at some providers |
| Providers with a standard price | 19 | 8 |
- Lower marks the lower price on a line; tie, equal prices. Other tiers and free rows are on the model pages.
- Promo: the provider calls the price promotional or discounted, so it is current now and can end without notice.
Report a wrong price or number (email to info@broadnet.ai, with this page's link filled in).
What do Qwen3.8 27B and Qwen3.6 27B cost for a real workload?
30 days at each model's cheapest standard price for the workload (presets of the cost calculator; cached input at the cached price where listed).
| Workload | Qwen3.8 27B | Qwen3.6 27B | Difference |
|---|---|---|---|
| Support chatbot 2,000 requests a day, 1,500 input tokens (60% cached), 300 output | USD 24.11 Chutes | USD 46.35 Chutes | USD 22.24 Qwen3.8 27B lower |
| RAG over documents 1,000 requests a day, 8,000 input tokens (30% cached), 500 output | USD 25.58 Chutes | USD 73.80 Chutes | USD 48.22 Qwen3.8 27B lower |
| Coding agent (one developer) 1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning | USD 129.00 Chutes | USD 285.75 Chutes | USD 156.75 Qwen3.8 27B lower |
Qwen3.8 27B is cheaper on all three workloads. The widest gap is on RAG over documents: the dearer model costs 2.88× as much as the cheaper one.
Does the cheaper model change with your input:output ratio?
Qwen3.8 27B is cheaper at every mix of input and output tokens, before cache discounts: its cheapest input price and its cheapest output price are both at or below Qwen3.6 27B's, so the input:output ratio never changes the answer.
| Input : output tokens | Qwen3.8 27B | Qwen3.6 27B |
|---|---|---|
| 1 : 3 (long answers) | USD 0.9472 lower Chutes | USD 1.535 Reka |
| 1 : 1 | USD 0.6445 lower Chutes | USD 1.12 Reka |
| 3 : 1 (the site’s blend) | USD 0.3417 lower Chutes | USD 0.6875 Chutes |
| 10 : 1 (retrieval, long prompts) | USD 0.1491 lower Chutes | USD 0.4091 Chutes |
| 30 : 1 (agents, big contexts) | USD 0.07806 lower Chutes | USD 0.3065 Chutes |
USD per 1M tokens at the cheapest standard provider for each mix, before cache discounts. Thinking tokens bill as output.
What do Qwen3.8 27B and Qwen3.6 27B cost per character in other languages?
Cost to send 1M characters at each model's cheapest non-promotional standard input price, with the Qwen3.8 27B tokenizer for Qwen3.8 27B (the same one for both, so the gap equals the price gap):
| Content | Qwen3.8 27B | Qwen3.6 27B |
|---|---|---|
| English | USD 0.0087 lower Chutes | USD 0.055 Chutes |
| Arabic MSA | USD 0.012 lower Chutes | USD 0.077 Chutes |
| Egyptian Arabic | USD 0.013 lower Chutes | USD 0.081 Chutes |
| Hindi | USD 0.016 lower Chutes | USD 0.106 Chutes |
| Chinese | USD 0.023 lower Chutes | USD 0.145 Chutes |
| Spanish | USD 0.0091 lower Chutes | USD 0.058 Chutes |
| Code | USD 0.011 lower Chutes | USD 0.070 Chutes |
From the real cost by language dataset (measured tokens per character on FLORES-200 and other texts). Your own text will differ.
Published results on the same benchmark
BroadNet did not run these benchmarks. BroadRouter is BroadNet's site. Each row pairs results from one source with the same setting for both models, as published; results from different sources or settings are never put side by side. One benchmark says little about your own workload.
| Benchmark | Qwen3.8 27B | Qwen3.6 27B | Source and setting |
|---|---|---|---|
| GPQA Diamond | 89.2 | 87.8 | Qwen model card: Qwen3.8 27B, Qwen3.6 27B, checked 2026-11-16 Setting not stated by the source |
| LiveCodeBench v6 | 90.3 | 83.9 | Qwen model card: Qwen3.8 27B, Qwen3.6 27B, checked 2026-11-16 Setting not stated by the source |
All published results for each model, with every setting: Qwen3.8 27B · Qwen3.6 27B.
Qwen3.8 27B on BroadNet Inference
BroadRouter is BroadNet's own site, so BroadNet's price is shown here for reference and is not part of the comparison above. BroadNet Inference serves Qwen3.8 27B as broadnet/qwen3.8-27b at USD 0.175 input, USD 0.0435 cached and USD 2.17 output per 1M tokens (USD 0.6737 blended 3:1; the cheapest third-party standard price above is USD 0.3417). On BroadNet it reads up to 262,144 tokens per request and accepts text and image input (BroadNet rate card).
Check it on your own traffic
- Qwen3.8 27B API pricing: all 19 providers and other tiers.
- Qwen3.6 27B API pricing: all 8 providers and other tiers.
- Cost calculator with Qwen3.8 27B and Qwen3.6 27B preloaded
- Qwen token counter
- Prompt caching calculator
More comparisons: DeepSeek V4.1 Flash vs Qwen3.8 27B · Gemma 4 31B vs Qwen3.8 27B · GLM 5.3 vs Qwen3.8 27B · gpt-oss-120b vs Qwen3.8 27B · Qwen3.8 27B vs Llama 4 Maverick · Qwen3.8 27B vs MiniMax M3 · Qwen3.8 27B vs DeepSeek V3.2 · Qwen3.8 27B vs MiniMax M3.1 Flash · Qwen3.6 27B vs Qwen3.5 27B · Qwen3.6 35B A3B vs Qwen3.6 27B · every Qwen3.8 27B comparison · every Qwen3.6 27B comparison · all 662 price comparisons
Questions
Is Qwen3.8 27B cheaper than Qwen3.6 27B?
Yes. On a blend of 3 input tokens to 1 output token, Qwen3.8 27B costs USD 0.3417 per 1M tokens and Qwen3.6 27B USD 0.6875, at each model's cheapest standard price. For input alone, Qwen3.8 27B is cheaper (USD 0.039 against USD 0.25 per 1M); for output alone, Qwen3.8 27B is (USD 1.25 against USD 1.95). This compares price only, not quality.
What do Qwen3.8 27B and Qwen3.6 27B cost per 1M tokens?
Qwen3.8 27B: from USD 0.039 input and USD 1.25 output at Chutes; Alibaba Cloud's own API charges USD 0.50 / USD 3; median of 19 standard prices USD 0.40 / USD 2.50. Qwen3.6 27B: from USD 0.25 input and USD 2 output at Chutes; Alibaba Cloud's own API charges USD 0.60 / USD 3.60; median of 8 standard prices USD 0.32 / USD 3.20. Each price is read from the provider's own pricing page; the source is linked in the table.
Which providers sell both Qwen3.8 27B and Qwen3.6 27B?
7 providers have a standard price for both: Chutes (USD 0.3417 and USD 0.6875 at 3:1), Reka (USD 0.775 and USD 0.705 at 3:1), DeepInfra (USD 0.5813 and USD 1.04 at 3:1), Featherless (USD 1.05 and USD 0.915 at 3:1), Venice (USD 1.138 and USD 1.056 at 3:1), Novita AI (USD 1.065 and USD 1.35 at 3:1), Alibaba Cloud (USD 1.125 and USD 1.35 at 3:1). The first is the lowest combined price. Buying each model from its own cheapest provider costs USD 0.3417 and USD 0.6875.
Which is cheaper for a coding agent or a chatbot, Qwen3.8 27B or Qwen3.6 27B?
For a coding agent (1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning): USD 129.00 a month on Qwen3.8 27B (Chutes) against USD 285.75 on Qwen3.6 27B (Chutes), a difference of USD 156.75. For a support chatbot (2,000 requests a day, 1,500 input tokens (60% cached), 300 output): USD 24.11 a month on Qwen3.8 27B (Chutes) against USD 46.35 on Qwen3.6 27B (Chutes), a difference of USD 22.24. Each uses the cheapest standard price for that workload, cached input included. Try your own numbers.
At what input:output ratio does the cheaper of Qwen3.8 27B and Qwen3.6 27B change?
Qwen3.8 27B is cheaper at every mix of input and output tokens, before cache discounts: its cheapest input price and its cheapest output price are both at or below Qwen3.6 27B's, so the input:output ratio never changes the answer.
Do Qwen3.8 27B and Qwen3.6 27B have cached input pricing?
Qwen3.8 27B from USD 0.0039 per 1M cached input tokens (Chutes; 14 of 19 standard prices list a discount). Qwen3.6 27B from USD 0.025 per 1M cached input tokens (Chutes; 3 of 8 standard prices list a discount). Cached tokens are billed at the input price where no cache price is published.
Which has the larger context window, Qwen3.8 27B or Qwen3.6 27B?
Qwen3.8 27B: up to 1,000,000 tokens, against 262,144 for Qwen3.6 27B, as stated on providers' pricing pages. Some providers offer less than the maximum; check the model pages before you pick the cheapest row for long prompts.
Can I use Qwen3.8 27B on BroadNet Inference?
BroadNet Inference serves Qwen3.8 27B as broadnet/qwen3.8-27b at USD 0.175 input and USD 2.17 output per 1M tokens (rate card synced 2026-11-14). BroadRouter is BroadNet's own site, so BroadNet's price is shown for reference and not ranked against the providers above.
Which scores higher on GPQA Diamond, Qwen3.8 27B or Qwen3.6 27B?
Per the Qwen model cards (Qwen3.8 27B, Qwen3.6 27B) (checked 2026-11-16), Qwen3.8 27B scores 89.2 and Qwen3.6 27B scores 87.8 on GPQA Diamond (setting not stated by the source). BroadNet did not run these benchmarks, and one benchmark says little about your own workload. Every result both models share, with sources.
Read as MarkdownAsk ChatGPTAsk ClaudeAsk Perplexity
Sources
- Chutes: pricing (checked 2026-11-15)
- Alibaba Cloud: own API pricing (checked 2026-11-14)
- Reka: pricing (checked 2026-11-16)
- DeepInfra: pricing (checked 2026-11-14)
- DeepInfra: pricing (checked 2026-11-14)
- Featherless: pricing (checked 2026-11-15)
- Featherless: pricing (checked 2026-11-15)
- Venice: pricing (checked 2026-11-15)
- Novita AI: pricing (checked 2026-11-14)
- Doubleword: pricing (checked 2026-11-16)
- Qwen model card, qwen3.8-27b (checked 2026-11-16)
- Qwen model card, qwen3.6-27b (checked 2026-11-16)
- BroadNet Inference rate card, pricing.json (synced 2026-11-14)