LLM Leaderboard
DeepSeek V4 Flash
by DeepSeek Reasoning
RunFree Score
59.1
#10 overall
Blended price
$0.11
per 1M tokens
Context
1.0M
tokens
Max output
66K
tokens
Output speed
—
tokens / sec
Latency (TTFT)
—
first token
Uptime
99.9%
last 24h
Providers
18
serving this model
Capability profile
Loading chart…
Compare DeepSeek V4 Flash
See it side by side with any other model — benchmarks, price and specs.
Benchmark results
| Benchmark | Category | Score | As of | Source |
|---|---|---|---|---|
| Terminal-Bench | agentic | 56.9% | Apr 24, 2026 | cite |
| LiveCodeBench | coding | 91.6% | Apr 24, 2026 | cite |
| SWE-Bench Verified | coding | 79% | Apr 24, 2026 | cite |
| Humanity's Last Exam | knowledge | 34.8% | Apr 24, 2026 | cite |
| MMLU-Pro | knowledge | 86.4% | Apr 24, 2026 | cite |
| GPQA Diamond | reasoning | 88.1% | Apr 24, 2026 | cite |
Every score links to its original source. See our methodology for how the RunFree Score is computed.
API pricing
- Input
- $0.09 / 1M
- Output
- $0.18 / 1M
- Cached input
- $0.02 / 1M
- Blended (3:1)
- $0.11 / 1M
Specifications
- Provider
- DeepSeek
- Context window
- 1.0M
- Max output
- 66K
- Reasoning model
- Yes
- Modality
- text
- Released
- Apr 24, 2026