Side-by-side comparison of DeepSeek-V4-Pro-0813, Qwen3.8-Max: pricing, capabilities, integrations and compliance — from verified HokAI records.
DeepSeek
Pick it if: DeepSeek-V4-Pro-0813 ships MIT-licensed open weights, with 49B parameters active per token and non-think, think-high, and think-max reasoning tiers callers pick per request. It suits cost-sensitive agentic coding teams willing to add an external safety layer, since independent 2026 red-teaming found refusal collapses under free-form prompting.
Its edge: SWE-bench Verified 80.6%, matching Gemini-3.1-Pro's agentic coding score, per DeepSeek V4 Pro GA coverage.
The catch: Text-only: no native vision, audio, or video input, despite pre-launch reporting that expected multimodal training.
Alibaba Cloud
Pick it if: Qwen3.8-Max launched GA on August 3, 2026 with 2.4 trillion total parameters (about 95 billion active) and a 1-million-token context window across a single flat pricing tier. It targets teams building large-context, multimodal agentic workloads on Alibaba Cloud who can tolerate its still-unpublished safety and training model card.
Its edge: Leads PaperBench at 93.0, ahead of GPT-5.6 Sol (90.5), Fable 5 (88.8) and Opus 4.8 (80.3).
The catch: No published safety or training model card as of the 2026-08-03 GA launch.
| Input / 1M tokens | $0.435 | $2 |
|---|---|---|
| Output / 1M tokens | $0.87 | $6 |
| Cached input / 1M | $0.004 | — |
| Blended cost (3:1) | $0.544 | $3 |
| Free tier | false | true |
| Pricing model | per-token | per-token |
| Strengths | SWE-bench Verified 80.6%, matching Gemini-3.1-Pro's agentic coding score, per DeepSeek V4 Pro GA coverage.; MMLU-Pro 87.5% and LiveCodeBench 93.5% pass@1, among the highest reported scores for an open-weight model.; Outputs at 80.0 tokens/s | Leads PaperBench at 93.0, ahead of GPT-5.6 Sol (90.5), Fable 5 (88.8) and Opus 4.8 (80.3).; 1-million-token context window (991.8K input non-thinking / 983.6K thinking) with a single flat pricing tier covering the full window.; $2.00 / $6.0 |
|---|---|---|
| Limitations | Text-only: no native vision, audio, or video input, despite pre-launch reporting that expected multimodal training.; FAR.AI's independent stress test found 98-100% jailbreak success across CBRN, cyber, and terrorism prompts using an unmodif | No published safety or training model card as of the 2026-08-03 GA launch.; Trails Fable 5 on Humanity's Last Exam (43.6 vs 53.3) and SWE-bench Pro (67.7 vs 80.0).; Open weights promised 'next week' at launch but not yet shipped; API-only f |
| Context window | 1M | 1M |
|---|---|---|
| Max output tokens | 384K | 131.1K |
| Input modalities | text; tool-calls | text; image; video |
| Output modalities | text; tool-calls | text |
| Capabilities | Tool use; Function calling; Structured output | Vision; Tool use; Video input |
| Reasoning modes | non-think; think-high; think-max | standard; thinking (preserved reasoning) |
| Long-context recall | — | unverified |
| Openness | open-source | proprietary |
Compare up to four at a time, or run Smart Match to get a ranked shortlist. Browse the full AI directory.