Skip to main content

Includes 694 hourly rounds up to 2026-11-16 from NIMStats by Mauro Druwel (source), measured with a different prompt and timeout. Our own probe adds a round every 30 minutes.

mistral

mistral-large-4-1t-a52b

Unstable

Rank 10 of 15 Last probed (22 min ago)

mistralai/mistral-large-4-1t-a52b answered 66.3% of probes on the NVIDIA NIM free API in the last 7 days, with a median time to first token of 23.3 s and 11.9 tok/s. It ranks 10 of 15 tracked models.

66.3% Uptime 63/95 probes
51/ 100 Score 0 round wins
23.3 s First token p95 165.8 s
38.1 s Full reply median
4.85 s Best reply fastest in range
11.9tok/s Throughput median

Model vs. median

Against the median of models that answered, last 7 days.

Uptime
66.3% 74.7%
8 pts below median
First token
23.3 s 14.8 s
1.6× worse than median
Full reply
38.1 s 22.6 s
1.7× worse than median
Throughput
11.9 tok/s 22.3 tok/s
1.9× worse than median

mistral-large-4-1t-a52b median of answering models

Error breakdown

32 failed of 95 probes, last 7 days.

FailureProbesShare of failures
Empty answer29 90.6%
Not served (404)2 6.3%
Timed out1 3.1%

Time to first token

Each probe. Lower is better.

0 ms125.0 s250.0 s375.0 s500.0 s11-13 09:0011-17 16:13 UTC
answered no answer

Throughput

Tokens per second after the first token, each probe. Higher is better.

0.0 tok/s25.0 tok/s50.0 tok/s75.0 tok/s100 tok/s11-13 09:0011-17 16:13 UTC
answered no answer

Availability heatmap

Last 7 days by UTC hour.

all answered some failed none answered not probed

Run history

Last 20 probes.

TimeResultFirst tokenFull replySpeed
OK5.30 s97.7 s2.2 tok/s
OK6.40 s36.3 s6.7 tok/s
OK342 ms44.6 s4.5 tok/s
OK372 ms26.1 s7.8 tok/s
OK344 ms8.66 s24.1 tok/s
OK2.17 s9.89 s25.9 tok/s
OK7.15 s57.1 s4.0 tok/s
OK2.33 s18.0 s12.8 tok/s
OK438 ms8.40 s25.1 tok/s
OK505 ms7.21 s29.8 tok/s
OK316 ms22.2 s9.1 tok/s
OK2.83 s39.3 s5.5 tok/s
OK30.8 s42.0 s11.9 tok/s
OK20.0 s22.0 s22.7 tok/s
OK293 ms15.5 s13.2 tok/s
OK229.5 s248.5 s2.0 tok/s
Empty answern/an/an/a
OK127.9 s151.8 s3.3 tok/s
Empty answern/an/an/a
OK20.0 s21.8 s22.9 tok/s

Call mistral-large-4-1t-a52b

The endpoint is OpenAI-compatible.

Create a free key at build.nvidia.com, then:

curl https://integrate.api.nvidia.com/v1/chat/completions \
  -H "Authorization: Bearer $NVIDIA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "mistralai/mistral-large-4-1t-a52b",
    "messages": [{"role": "user", "content": "Hello"}],
    "max_tokens": 256,
    "stream": true
  }'

Other models answering

Top of the leaderboard right now.