DeepSeek V4.1 Flash

deepseek/deepseek-v4.1-flash

openrouter1 build benchmarkedbest build deepseek215 tok/s
best build

Scores at a glance

overall60
Tool Calling7819/27
Structured Output5214/27
RAG / Retrieval QA6317/27
Context Recall10018/18
Coding4111/27
Reasoning & Math3911/28
Instruction Following4613/28
Classification5516/29
Summarization7019/27
0255075100

0–100 per task · n/m = cases passed (scores credit partial passes) · overall = mean across tasks

best build
config
deployment

deepseek

quantization
harness
openrouter
context
1.0M
memory
throughput
215 tok/s
cost / run
$0.7017
provider
openrouter
run
2026-09-10
run this buildOpenRouterhosted
curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"deepseek/deepseek-v4.1-flash","messages":[{"role":"user","content":"Hello"}]}'

Hosted API — not self-hostable. This is the call the benchmark made; try it live in the Playground.

record
per-test results

Full run record

Per-test scores, full transcripts, and the reproduce recipe.

open the run record →