Mercury 2.5 Preview vs Qwen3.8 Max Prime
Inception Labs · US | Alibaba · China · Updated June 2026
Quick verdict
Pick Mercury 2.5 Preview for very high inference throughput (1,107 tok/s) via diffusion-based parallel token generation or coding accuracy (95.7%, 91st percentile among cost-optimized models). Pick Qwen3.8 Max Prime for same 2.4t-parameter moe architecture and full capability set as qwen3.8 max, repackaged for higher-throughput serving or alibaba advertises 1.5-2x higher output throughput than standard qwen3.8 max for latency-sensitive workloads. On a tight budget at scale, Mercury 2.5 Preview is the value pick.
Mercury 2.5 Preview (Inception Labs, US) and Qwen3.8 Max Prime (Alibaba, China) line up two different AI ecosystems against each other — a comparison that is as much about cost philosophy and openness as raw capability. Mercury 2.5 Preview is inception Labs' August 31, 2026 diffusion-based language model preview, refining tokens in parallel for roughly 10x the throughput of comparable autoregressive models at cost-optimized-tier quality. Qwen3.8 Max Prime is alibaba's September 23, 2026 high-throughput variant of Qwen3.8 Max — identical 2.4T-parameter architecture at double the price, with throughput gains public testing hasn't yet confirmed. They diverge most on price and context window — each quantified below from the models' real specs.
Key differences at a glance
- ▸Price: Mercury 2.5 Preview is about 100× cheaper on input ($0.04/$0.15 per 1M tokens vs $4/$12 per 1M tokens) — a large enough gap that at scale it can be the single biggest line item in the decision.
- ▸Context window: Qwen3.8 Max Prime holds 4× more — 1M tokens (~1,573 pages) vs 260K tokens (~390 pages). But effective recall usually fades long before the advertised ceiling, so the bigger number only helps if the model reasons over it.
- ▸Recency: Qwen3.8 Max Prime is the newer model by about 23 days (released September 23, 2026), usually meaning fresher training data and capabilities.
- ▸Ecosystem: this is a US-vs-China matchup — they differ in pricing philosophy, data-residency options, and tooling ecosystems, not only benchmarks.
Side-by-side specs
| Spec | Mercury 2.5 Preview | Qwen3.8 Max Prime |
|---|---|---|
| Provider | Inception Labs (US) | Alibaba (China) |
| Released | August 31, 2026 | September 23, 2026 |
| Context window | 260K tokens (~390 pages) | 1M tokens (~1,573 pages) |
| Price (in/out) | $0.04/$0.15 per 1M tokens | $4/$12 per 1M tokens |
| Open weight? | No — API only | No — API only |
| Modalities | text | text, image, video, code |
| SWE-Bench Verified | Not published | Not published |
| MRCR v2 @ 1M | Not published | Not published |
Who wins what
Very high inference throughput (1,107 tok/s) via diffusion-based parallel token generation
Mercury 2.5 Preview
Qwen3.8 Max Prime is comparatively weak here — priced exactly double standard Qwen3.8 Max ($4/$12 vs $2/$6) for a throughput claim the only public measurements available don't show as measurably faster
Coding accuracy (95.7%, 91st percentile among cost-optimized models)
Mercury 2.5 Preview
At $0.04/$0.15 per 1M tokens it undercuts Qwen3.8 Max Prime ($4/$12 per 1M tokens), and that gap compounds at volume.
Mathematics accuracy (97.0%, 97th percentile)
Mercury 2.5 Preview
Inception Labs' August 31, 2026 diffusion-based language model preview, refining tokens in parallel for roughly 10x the throughput of comparable autoregressive models at cost-optimized-tier quality — and it runs cheaper at $0.04/$0.15 per 1M tokens.
Same 2.4T-parameter MoE architecture and full capability set as Qwen3.8 Max, repackaged for higher-throughput serving
Qwen3.8 Max Prime
Alibaba's September 23, 2026 high-throughput variant of Qwen3.8 Max — identical 2.4T-parameter architecture at double the price, with throughput gains public testing hasn't yet confirmed — and it carries the larger 1M tokens context.
Alibaba advertises 1.5-2x higher output throughput than standard Qwen3.8 Max for latency-sensitive workloads
Qwen3.8 Max Prime
Alibaba's September 23, 2026 high-throughput variant of Qwen3.8 Max — identical 2.4T-parameter architecture at double the price, with throughput gains public testing hasn't yet confirmed — and it is the newer of the two.
1M-token context window with up to 131,072 output tokens, reasoning enabled by default
Qwen3.8 Max Prime
Its 1M tokens window holds about 4× more than Mercury 2.5 Preview's 260K tokens in a single prompt.
Lowest cost at scale
Mercury 2.5 Preview
At $0.04/$0.15 per 1M tokens, it is the cheaper of the two — the gap dominates the bill on high-volume workloads.
Largest single-prompt input
Qwen3.8 Max Prime
Its 1M tokens window is about 4× larger than Mercury 2.5 Preview's 260K tokens, fitting roughly 1,573 pages in one prompt.
Which should you pick?
A cost-sensitive startup shipping high volume
→ Mercury 2.5 Preview
At $0.04/$0.15 per 1M tokens it undercuts Qwen3.8 Max Prime, and on millions of tokens that margin decides the monthly bill.
Someone analysing very long documents or codebases
→ Qwen3.8 Max Prime
Larger 1M tokens window fits more in one prompt.
Anyone whose priority is very high inference throughput (1,107 tok/s) via diffusion-based parallel token generation
→ Mercury 2.5 Preview
It is specifically built for that.
Anyone whose priority is same 2.4t-parameter moe architecture and full capability set as qwen3.8 max, repackaged for higher-throughput serving
→ Qwen3.8 Max Prime
That is its strongest area.
An enterprise with regional data-residency rules
→ Mercury 2.5 Preview or Qwen3.8 Max Prime
Origin (US vs China) affects where data is processed and which compliance regime applies — check the provider's terms for your region.
Mercury 2.5 Preview: where it fits
Inception Labs' August 31, 2026 diffusion-based language model preview, refining tokens in parallel for roughly 10x the throughput of comparable autoregressive models at cost-optimized-tier quality. Released August 31, 2026 by Inception Labs, it is built for very high inference throughput (1,107 tok/s) via diffusion-based parallel token generation, coding accuracy (95.7%, 91st percentile among cost-optimized models), mathematics accuracy (97.0%, 97th percentile), and tunable reasoning levels with parallel tool calls and schema-aligned JSON output.
Its trade-offs are real: 260K context window is far shorter than frontier 1M-token models, no vision or audio modalities, positioned only against cost-optimized models (GPT-5.6 Luna Low, Gemini 3.5 Flash-Lite), not frontier-class, and current $0.04/$0.15 pricing includes a limited-time launch promotion on OpenRouter (list price is $0.20/$0.75, may rise after Sept 8, 2026). At $0.04 in / $0.15 out per million tokens, it sits in the budget price band.
Qwen3.8 Max Prime: where it fits
Alibaba's September 23, 2026 high-throughput variant of Qwen3.8 Max — identical 2.4T-parameter architecture at double the price, with throughput gains public testing hasn't yet confirmed. Released September 23, 2026 by Alibaba, it is built for same 2.4T-parameter MoE architecture and full capability set as Qwen3.8 Max, repackaged for higher-throughput serving, alibaba advertises 1.5-2x higher output throughput than standard Qwen3.8 Max for latency-sensitive workloads, 1M-token context window with up to 131,072 output tokens, reasoning enabled by default, and accepts text, image, and video input alongside text output, inherited unchanged from the base model.
Its trade-offs: priced exactly double standard Qwen3.8 Max ($4/$12 vs $2/$6) for a throughput claim the only public measurements available don't show as measurably faster, no official Alibaba announcement — Alibaba Cloud's Model Studio documentation still lists only qwen3.8-max and its September 2 snapshot, suggesting minimal official backing, like the base model, open weights have been promised but not shipped — closed API only, and purely a speed tier, not a capability upgrade — same benchmark ceiling as the cheaper base model. At $4 in / $12 out per million tokens, it sits in the mid price band.
The bottom line for this matchup
This is less "which is smarter" and more "which ecosystem fits." Mercury 2.5 Preview (US) and Qwen3.8 Max Prime (China) differ on pricing philosophy, data-residency, and tooling as much as on raw scores. Mercury 2.5 Preview is the cheaper option, which matters at volume. The pragmatic move is to run one real task through both and judge the outputs against your own constraints — including where your data is allowed to be processed.
Want both Mercury 2.5 Preview and Qwen3.8 Max Prime without two subscriptions? LumiChats gives you these plus 40+ models under one ₹69/day pass (about $1/day) — draft with one, cross-check with the other.
See pricingFrequently asked questions
Is Mercury 2.5 Preview or Qwen3.8 Max Prime better for coding?
Public SWE-Bench figures are not available for either model, so the honest test is your own repository — run an identical real bug through both. By design, Mercury 2.5 Preview leans toward very high inference throughput (1,107 tok/s) via diffusion-based parallel token generation while Qwen3.8 Max Prime leans toward same 2.4t-parameter moe architecture and full capability set as qwen3.8 max, repackaged for higher-throughput serving, and that positioning usually predicts which feels better on your codebase.
Which is cheaper, Mercury 2.5 Preview or Qwen3.8 Max Prime?
Mercury 2.5 Preview is cheaper — $0.04/$0.15 per 1M tokens vs $4/$12 per 1M tokens, roughly 100× apart on input.
Which has the bigger context window?
Qwen3.8 Max Prime — 1M tokens vs 260K tokens, about 4× larger. Useful only if the model actually reasons over the full window, which not all do.
Can I use both Mercury 2.5 Preview and Qwen3.8 Max Prime together?
Yes — a multi-model platform like LumiChats gives you Mercury 2.5 Preview, Qwen3.8 Max Prime and 40+ others under one ₹69/day pass (about $1/day), so you can draft with one and cross-check with the other instead of buying two subscriptions.
Which is newer, Mercury 2.5 Preview or Qwen3.8 Max Prime?
Qwen3.8 Max Prime — released September 23, 2026, about 23 days after Mercury 2.5 Preview.
Related comparisons
Specifications and benchmarks reflect publicly reported figures as of June 2026 and may change as providers release updates. Always verify on your own workload.