Parasail: Models Intelligence, Performance & Price

Parasail
Parasail

This analysis is intended to support you in choosing the best model provided by Parasail for your use-case.

Most Intelligent

#1
Kimi K3.1 ProKimi K3.1 Pro
46
#2
DeepSeek V4.1 Flash 1017 (max)DeepSeek V4.1 Flash 1017 (max)
33
#3
GLM-5.3 (max)GLM-5.3 (max)
32
#4
MiniMax-M3.1-FlashMiniMax-M3.1-Flash
29
#5
GLM-5.3-FlashGLM-5.3-Flash
28

Intelligence Index

Total 31 models

Lowest Time per Task

#1
gpt-oss-120b (high)gpt-oss-120b (high)
6.8 min
#2
GLM-5.3-FlashGLM-5.3-Flash
7.2 min
#3
GLM-5.3 (max)GLM-5.3 (max)
9.5 min
#4
DeepSeek V4.1 Flash 1017 (max)DeepSeek V4.1 Flash 1017 (max)
10.2 min
#5
Kimi K3.1 ProKimi K3.1 Pro
10.5 min

Time per Intelligence Index task (minutes)

Total 31 models

Lowest Price

#1
Gemma 3 4B (FP8)Gemma 3 4B (FP8)
$0.05
#2
Gemma 3 27BGemma 3 27B
$0.09
#3
MiniMax-M3.1-FlashMiniMax-M3.1-Flash
$0.10
#4
GLM-5.3-FlashGLM-5.3-Flash
$0.10
#5
Gemma 4 26B A4BGemma 4 26B A4B
$0.10

Blended price (per 1M tokens)

Total 31 models

Parasail offers 31 models, each with different intelligence, performance, and pricing characteristics. Below is a comparison of the key metrics across models.

  • For intelligence, the top models on Parasail are Kimi K3.1 Pro (46), DeepSeek V4.1 Flash 1017 (max) (33), and GLM-5.3 (max) (32).
  • For time per task, gpt-oss-120b (high) (6.8 min), GLM-5.3-Flash (7.2 min), and GLM-5.3 (max) (9.5 min) finish an Intelligence Index task fastest on Parasail.
  • For latency, Qwen3 Next 80B A3B (0.99s), Qwen3 Coder Next (FP8) (1.04s), and Qwen3.6 35B A3B (non-reasoning) (FP8) (1.13s) offer the lowest time to first answer token.
  • For pricing, Gemma 3 4B (FP8) ($0.05), Gemma 3 27B ($0.09), and MiniMax-M3.1-Flash ($0.10) offer the lowest blended prices per 1M tokens. Prices vary up to 2.2x across models.
  • For context window size, Kimi K3.1 Pro (1M), Llama 4 Maverick (FP8) (1M), and DeepSeek V4 Flash (max) (FP8) (1M) support the largest context windows on Parasail.
Artificial Analysis Intelligence Index · Higher is better
Inference time (minutes) per Intelligence Index task · Lower is better
USD per 1M tokens (blended) · Lower is better

Intelligence

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v5.0.1 incorporates 9 evaluations: AA-Briefcase v1.1, AutomationBench-AA, Terminal-Bench 4.0, FrontierCode-AA 2.0, AA-Omniscience v1.1, GDP.pdf, HLE-Diamond, CritPt v2 (Hard Subset), and Terminal-Bench-Science 0.1

Evaluations

Evaluations measured independently by Artificial Analysis · AA-PersonalityBench scores are percentile ranks from 1 to 100
See more

Agentic knowledge work, Elo

Agentic SaaS workflows

Agentic scientific research workflows in a terminal

Agentic software engineering in open-source repositories

Reasoning & knowledge

Professional document reasoning, All-pass

Agentic coding & terminal use

Physics reasoning

Quantitative analysis on spreadsheets & documents

Legal agentic work, Hallucination-Gated All-Pass Rate

Agentic business operations

Agentic real-world work tasks, Elo

Agentic interactive 3D experiences, Elo

Product and brand videos designed in code, Elo

Kubernetes incident root-cause analysis

Medical long context reasoning

Visual reasoning

Rich, varied, well-crafted language

Well-structured, scannable, answer-first

Truthful, acknowledges uncertainty, non-sycophantic

Use of filler and generic language

Intelligence Index vs. Price

Blended at 7:2:1 (cache-input-output) · USD per 1M tokens (blended)
Most attractive quadrant
Pareto line

Pricing

Cost per Intelligence Index Task

Weighted average cost (USD) per Intelligence Index task · Segmented by token type · Lower is better

Intelligence Index vs. Price

Blended at 7:2:1 (cache-input-output) · USD per 1M tokens (blended)
Most attractive quadrant
Pareto line

Speed

Average of the single-request and multi-turn agentic workloads measured

Time per Intelligence Index Task

Output tokens divided by end-to-end output speed (minutes per Intelligence Index task) · Lower is better

Quality

Endpoint Accuracy Index

Each endpoint's accuracy as a % of its model's reference endpoint · Color shows whether the gap is statistically significant

Latency

Average of the single-request and multi-turn agentic workloads measured

Time to First Answer Token

Seconds until the first answer token after any reasoning · Average of the single-request and multi-turn agentic workloads measured · Lower is better

Reliability

Share of requests the endpoint failed, across all the single-request and multi-turn agentic workloads measured

Endpoint Error Rate

Share of requests the endpoint failed with an error, a rate limit or a timeout · Across all the single-request and multi-turn agentic workloads measured · Lower is better

Caching

Cache Hit Rate

Median across multi-turn agentic session runs in the last four weeks, updated Oct 26, 2026 · Higher is better

Cost per Task vs. Cache Hit Rate

Weighted average cost (USD) per Intelligence Index task · Cache hit rate: median of the last four weeks, updated Oct 26, 2026
Most attractive quadrant
Pareto line

Further Analysis
Kimi logo
Kimi K3.1 Pro
46
$4.22
13.3
64.80
0.8%
1M
Open
DeepSeek logo
DeepSeek V4.1 Flash 1017 (max)
33
$0.52
7.6
29.69
0.0%
1M
Open
Z AI logo
GLM-5.3 (max)
32
$4.92
12.6
41.56
3.1%
1M
Open
MiniMax logo
MiniMax-M3.1-Flash
29
$0.44
14.2
29.19
3.0%
1M
Open
Z AI logo
GLM-5.3-Flash
28
$0.64
8.8
19.49
10.2%
1M
Open
Alibaba logo
Qwen3.8 27B (xhigh) (FP8)
20
$1.78
23.2
70.73
2.3%
262k
Open
MiniMax logo
MiniMax-M3 (MXFP8)
19
$1.17
14.5
8.67
9.6%
1M
Open
Google logo
Gemma 4 31B
4
--
--
19.58
3.6%
262k
Open
OpenAI logo
gpt-oss-120b (high)
4
$0.17
7.1
15.92
1.7%
131k
Open
Kimi logo
Kimi K2.6
--
--
--
93.80
41.2%
262k
Open
Alibaba logo
Qwen3.5 397B A17B
--
--
--
36.30
3.6%
262k
Open
Kimi logo
Kimi K2.6 (non-reasoning) (INT4)
--
--
--
2.64
8.1%
262k
Open
Google logo
Gemma 3 4B (FP8)
--
--
--
2.89
13.9%
131k
Open
Alibaba logo
Qwen3 Next 80B A3B
--
--
--
1.01
1.8%
262k
Open
Google logo
Gemma 3 27B
--
--
--
12.86
0.0%
131k
Open
Alibaba logo
Qwen3 235B 2507
--
--
--
2.58
0.0%
131k
Open
Alibaba logo
Qwen3.6 35B A3B (non-reasoning) (FP8)
--
--
--
1.16
1.6%
262k
Open
Google logo
Gemma 4 26B A4B
--
--
--
13.40
1.5%
256k
Open
Google logo
Gemma 4 26B A4B (non-reasoning)
--
--
--
1.68
1.4%
262k
Open
Alibaba logo
Qwen3.6 35B A3B
--
--
--
25.81
2.6%
262k
Open
DeepSeek logo
DeepSeek V4 Flash (high) (FP8)
--
--
--
--
--
1M
Open
Meta logo
Llama 4 Maverick (FP8)
--
--
--
1.21
5.3%
1M
Open
Meta logo
Llama 3.3 70B (FP8)
--
--
--
7.40
5.2%
131k
Open
OpenAI logo
gpt-oss-120b (low)
--
--
--
3.27
4.5%
131k
Open
Google logo
Gemma 4 31B (non-reasoning)
--
--
--
2.51
1.3%
262k
Open
Alibaba logo
Qwen3 VL 235B A22B (FP8)
--
--
--
3.18
8.3%
131k
Open
DeepSeek logo
DeepSeek V4 Flash (max) (FP8)
--
--
--
8.50
5.7%
1M
Open
DeepSeek logo
DeepSeek V4.1 Flash (max)
--
--
--
28.52
0.0%
1M
Open
Z AI logo
GLM-5.2 (max) (NVFP4)
--
--
--
16.85
5.5%
1M
Open
Kimi logo
Kimi K3 (max)
--
--
--
29.78
0.0%
1M
Open
Alibaba logo
Qwen3 Coder Next (FP8)
--
--
--
1.03
6.1%
262k
Open
DeepSeek logo
DeepSeek V4 Flash 0731 (max)
--
--
--
14.89
0.0%
1M
Open

Key definitions

Frequently Asked Questions

Common questions about Parasail