# DeepSeek-V4-Flash-0731

DeepSeek-V4-Flash-0731 is an AI model released by DeepSeek on Jul 31 2026. It has 284B parameters, a 1M context window and open weight. Tracked results include 39% on BullshitBench v2, 79% on SWE-Bench Verified and 34.8% on Humanity's Last Exam (with tools).

## Facts

| Field | Value |
| --- | --- |
| Model | DeepSeek-V4-Flash-0731 |
| Developer | DeepSeek |
| Release date | Friday, Jul 31 2026 |
| Licensing | Open Weight |
| Parameters | 284B |
| Context window | 1M |

## Tracked benchmark scores

| Benchmark | Score | Source | What it measures |
| --- | --- | --- | --- |
| BullshitBench v2 | 39% | [BullshitBench](https://github.com/petergpt/bullshit-benchmark) | Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better. |
| SWE-Bench Verified | 79% | [BenchLM](https://benchlm.ai), retrieved 2026-08-31 | Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. |
| Terminal-Bench 2.1 | 82.7% | Lab | Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Higher is better. |
| Toolathlon-Verified | 70.3% | Lab | Tests how well the AI uses everyday personal tools and apps to get things done — a human-checked version of Toolathlon. Higher is better. |
| BrowseComp | 73.2% | [BenchLM](https://benchlm.ai), retrieved 2026-08-31 | Can the AI browse the web and track down hard-to-find answers? Higher is better. |
| CyberGym | 76.7% | Lab | Tests the AI on cybersecurity challenges — finding and exploiting software weaknesses inside a safe sandbox. Higher is better. |
| Humanity's Last Exam (with tools) | 34.8% | [BenchLM](https://benchlm.ai), retrieved 2026-08-31 | Humanity's Last Exam — extremely hard expert questions across many subjects. “With tools” means the AI is allowed to search the web or run code while answering. Higher is better. |
| ARC-AGI-2 | 61.4% | [BenchLM](https://benchlm.ai), retrieved 2026-09-28 | Puzzle-style tests of abstract reasoning and pattern-finding — the kind of thing people find easy but AIs often struggle with. Higher is better. |
| AutomationBench | 25.1% | Lab | Tests whether the AI can run real multi-step business workflows — the kind of end-to-end office processes companies want to automate — from start to finish. Higher is better. |

## About DeepSeek-V4-Flash-0731

DeepSeek V4 Flash 0731 was the updated V4 Flash checkpoint that DeepSeek’s model card described as superseding the preview version. DeepSeek announced it on July 31, 2026 as the official V4-Flash API, in public beta under the unchanged deepseek-v4-flash name, with native Responses API support adapted for Codex. The card emphasized agentic tasks including software engineering and security work. It documented a model structure shared with the V4-Flash-DSpark build, including its speculative-decoding module.

DeepSeek published this checkpoint’s weights under the MIT License. The model card reported 82.7% on Terminal-Bench 2.1 using a minimal DeepSeek Harness at maximum reasoning effort; that is a result for the stated model-and-agent setup. The dated checkpoint name distinguishes it from both the April V4 Flash preview and the later V4.1 Flash architecture.

## Questions and answers

### When was DeepSeek-V4-Flash-0731 released?

DeepSeek-V4-Flash-0731 was released by DeepSeek on Friday, Jul 31 2026.

### Who made DeepSeek-V4-Flash-0731?

DeepSeek-V4-Flash-0731 was built by DeepSeek. Chinese AI lab known for efficient, open-weight models. Gained attention for strong performance at lower cost.

### What benchmark scores did DeepSeek-V4-Flash-0731 get?

DeepSeek-V4-Flash-0731 reports 9 tracked benchmark scores — BullshitBench v2: 39%; SWE-Bench Verified: 79%; Terminal-Bench 2.1: 82.7%; Toolathlon-Verified: 70.3%; BrowseComp: 73.2%; CyberGym: 76.7%; Humanity's Last Exam (with tools): 34.8%; ARC-AGI-2: 61.4%; AutomationBench: 25.1%. Tracked scores may come from lab reports or independent benchmarks; source details accompany the benchmark data.

### What is the context window of DeepSeek-V4-Flash-0731?

DeepSeek-V4-Flash-0731 has a context window of 1M. Providers may specify separate input and output token limits.

### How many parameters does DeepSeek-V4-Flash-0731 have?

DeepSeek-V4-Flash-0731 is reported at 284B parameters.

### Is DeepSeek-V4-Flash-0731 open source?

Partly. DeepSeek-V4-Flash-0731 is an open-weight model: the trained weights are free to download and run locally or on your own infrastructure, but the training data and code are not fully released and the license may restrict some commercial uses. It is not open source in the strict sense.

### What came before and after DeepSeek-V4-Flash-0731?

DeepSeek's previous tracked release was DeepSeek-V4-Flash on Apr 24 2026, 98 days earlier. It was followed by DeepSeek-V4-Pro-0813 on Aug 13 2026.


---

Canonical page: https://aireleasetracker.com/model/deepseek/deepseek-v4-flash-0731
Site index: https://aireleasetracker.com/llms.txt
Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.