$ diff deepseek-v4-flash inkling
DeepSeek V4-Flash vs Inkling
Values come from DeepSeek V4-Flash and Inkling, where each is cited to its source. This page states no benchmark result and ranks neither model — it puts two published specifications next to each other. Where a lab has not published a figure, the row says so rather than guessing.
DeepSeekDeepSeek V4-Flash
- context
- 1.0M
- weights
- open
- $/M in
- $0.14
- $/M out
- $0.28
- context
- 1M
- weights
- open
- $/M in
- —
- $/M out
- —
What actually differs
- Context window
- DeepSeek V4-Flash takes 1.0M against 1M — modestly more room in a single request.
- Weights
- Both publish weights — DeepSeek V4-Flash under MIT (open-weight), Inkling under Apache 2.0 (open-weight). Check the licences rather than assuming they permit the same commercial use.
- Recency
- DeepSeek V4-Flash shipped 16 days after Inkling (2026-07-31 vs 2026-07-15).
Full spec
| Attribute | DeepSeek V4-Flash | Inkling |
|---|---|---|
| Developer | DeepSeek | Thinking Machines |
| Released | 2026-07-31 | 2026-07-15 |
| Context window | 1,048,576 (1M) — OpenRouter catalogue `deepseek/deepseek-v4-flash`, read 2026-08-01; DeepSeek's own launch material did not state it | 1M |
| Pricing | $0.14/M input (cache miss) · $0.0028/M input (cache hit) · $0.28/M output — see Conflicting Reports | not recorded |
| License | MIT (open-weight) | Apache 2.0 (open-weight) |
| Availability | DeepSeek API (Responses format, Codex-adapted), Hugging Face (`deepseek-ai/DeepSeek-V4-Flash-0731`) | Hugging Face (BF16 and NVFP4), Tinker (fine-tuning), SGLang, vLLM, llama.cpp, Unsloth |