eva-unit-01 · Open weights
EVA-LLaMA-3.33-70B-v0.1
A RP/storywriting specialist model, full-parameter finetune of Llama-3.3-70B-Instruct on mixture of synthetic and natural data. It uses Celeste 70B 0.1 data mixture, greatly expanding it to improve versatility, creativity and flavor of the resulting model.
- Input
- Text
- Output
- Text
- Discovery tier
- Open weights
Pricing
USD per 1M tokens- Input
- $2.01
- Output
- $2.01
Reference prices as published by the model provider. API example
Specifications
- Context window
- 32.8K
- Max output
- —
- Average output (7 days)
- 459 tokens
- Longer than 45% of 468 models
Side by side
EVA-LLaMA-3.33-70B-v0.1 compared
Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.
| EVA-LLaMA-3.33-70B-v0.1eva-unit-01This model | Mixtral 8x22BMistralRemove | |
|---|---|---|
| Provider | eva-unit-01 | Mistral |
| Discovery tier | Open weights | Open weights |
| Added | Sep 25, 2025 | Dec 11, 2025 |
| Limits | ||
| Context window | 32.77K | 65.54KBest |
| Max output | — | 52.43K |
| Average output | 459 tokens | 292 tokens |
| Pricing, USD per 1M tokens | ||
| Input | $2.01 | $2.00Best |
| Output | $2.01Best | $6.00 |
| Cache read | — | $0.20 |
| Performance | ||
| Throughput | 17.5 tok/s | — |
| Time to first token | 8.6 s | — |
| Request success (24h) | 25.5% | — |
| Benchmark indexes | ||
| Intelligence Index | — | 5.7 |
| Capabilities | ||
| Reasoning effort | — | — |
Public evaluations
Benchmarks
Independent evaluation results, with how each score ranks among the models in the Noumen catalog.
- EVA-LLaMA-3.33-70B-v0.1
- Mixtral 8x22B
Intelligence Index
- EVA-LLaMA-3.33-70B-v0.1—
- Mixtral 8x22B5.7
Reasoning and knowledge
Humanity’s Last Exam
Expert-level questions across academic fields
- EVA-LLaMA-3.33-70B-v0.1: —
- Mixtral 8x22B: 4.0%
GPQA Diamond
Graduate-level science questions
- EVA-LLaMA-3.33-70B-v0.1: —
- Mixtral 8x22B: 33.2%
MMLU-Pro
Multi-subject knowledge and reasoning
- EVA-LLaMA-3.33-70B-v0.1: —
- Mixtral 8x22B: 53.7%
Coding
LiveCodeBench
Recent competitive programming problems
- EVA-LLaMA-3.33-70B-v0.1: —
- Mixtral 8x22B: 14.8%
Math
AIME
Competition mathematics
- EVA-LLaMA-3.33-70B-v0.1: —
- Mixtral 8x22B: 0.0%
MATH-500
Competition math problems
- EVA-LLaMA-3.33-70B-v0.1: —
- Mixtral 8x22B: 54.5%
Scores from Artificial Analysis, updated Oct 26, 2026.
Live routing data
Performance and reliability
Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.
- Throughput
- 17.5 tok/s
- Faster than 11% of 254 models
- Time to first token
- 8.6 s
- Quicker than 14% of 254 models
- Prompt cache hits
- 0.0%
- Request success (24h)
- 25.5%
Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.
Developers
Use it through the API
Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.
curl https://www.noumen.xyz/api/v1/chat/completions \
-H "Authorization: Bearer $NOUMEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1",
"messages": [{ "role": "user", "content": "Explain this stack trace." }]
}'import os
from openai import OpenAI
client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
model="EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1",
messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)eva-unit-01
Related models
Other eva-unit-01 models, starting with the same family.