Menu

logo
Artificial Analysis
HOME
 AI TrendsNEW
logo

Fireworks: Models Intelligence, Performance & Price

Analysis of Fireworks's models across key metrics including quality, price, output speed, latency, context window & more. This analysis is intended to support you in choosing the best model provided by Fireworks for your use-case. For more details including relating to our methodology, see our FAQs. Models analyzed: Llama 3.3 70B, Llama 3.1 405B, Llama 4 Maverick (Base), Llama 4 Scout (Base), Mixtral 8x22B, DeepSeek V3, DeepSeek R1 (May '25) Fast, Qwen3 30B A3B (Reasoning), QwQ-32B, Qwen3 235B (Reasoning), Yi-Large, Llama 3.1 70B, Llama 3.1 8B, DeepSeek R1 (Jan '25) (Fast), DeepSeek V3 (Dec '24), and Qwen2.5 72B.
Link:

Fireworks Model Comparison Summary

Intelligence:DeepSeek R1 (May '25) Fast logo DeepSeek R1 (May '25) Fast and Qwen3 235B (Reasoning) logo Qwen3 235B (Reasoning) are the highest quality models offered by Fireworks, followed by DeepSeek R1 (Jan '25) (Fast) logo DeepSeek R1 (Jan '25) (Fast), QwQ-32B logo QwQ-32B & Qwen3 30B A3B (Reasoning) logo Qwen3 30B A3B (Reasoning).Output Speed (tokens/s):Llama 3.1 8B logo Llama 3.1 8B (299 t/s) and DeepSeek V3 logo DeepSeek V3 (271 t/s) are the fastest models offered by Fireworks, followed by DeepSeek R1 (May '25) Fast logo DeepSeek R1 (May '25) Fast, DeepSeek R1 (Jan '25) (Fast) logo DeepSeek R1 (Jan '25) (Fast) & QwQ-32B logo QwQ-32B.Latency (seconds):Llama 3.1 8B logo Llama 3.1 8B (0.22s) and  Mixtral 8x22B logo Mixtral 8x22B (0.33s) are the lowest latency models offered by Fireworks, followed by Qwen2.5 72B logo Qwen2.5 72B, Yi-Large logo Yi-Large & Llama 3.1 70B logo Llama 3.1 70B.Blended Price ($/M tokens):Qwen3 235B (Reasoning) logo Qwen3 235B (Reasoning) ($0.10) and Llama 3.1 8B logo Llama 3.1 8B ($0.20) are the cheapest models offered by Fireworks, followed by Llama 4 Scout (Base) logo Llama 4 Scout (Base), Llama 4 Maverick (Base) logo Llama 4 Maverick (Base) & Llama 3.3 70B logo Llama 3.3 70B.Context Window Size:Llama 4 Maverick (Base) logo Llama 4 Maverick (Base) (1m) and Llama 4 Scout (Base) logo Llama 4 Scout (Base) (1m) are the largest context window models offered by Fireworks, followed by DeepSeek R1 (May '25) Fast logo DeepSeek R1 (May '25) Fast, DeepSeek R1 (Jan '25) (Fast) logo DeepSeek R1 (Jan '25) (Fast) & DeepSeek V3 logo DeepSeek V3.

Highlights

Intelligence
Artificial Analysis Intelligence Index; Higher is better
Speed
Output Tokens per Second; Higher is better
Price
USD per 1M Tokens; Lower is better
Parallel Queries:
Prompt Length:
Features
Model Intelligence
Price
Output tokens/s
Latency
End-to-End Response Time
Further
Analysis
Fireworks Fast logo
DeepSeek logo
DeepSeek R1 (May '25) Fast
164k
68
$4.25
253.4
0.45
10.32
7.89
Fireworks logo
Alibaba logo
Qwen3 235B (Reasoning)
128k
62
$0.10
97.1
0.50
26.24
20.59
Fireworks (Fast) logo
DeepSeek logo
DeepSeek R1 (Jan '25) (Fast)
164k
60
$4.25
238.8
0.49
12.41
9.83
Fireworks logo
Alibaba logo
QwQ-32B
131k
58
$0.90
179.4
0.43
17.10
13.89
Fireworks logo
Alibaba logo
Qwen3 30B A3B (Reasoning)
131k
56
$0.90
171.6
0.40
14.97
11.65
Fireworks logo
DeepSeek logo
DeepSeek V3
160k
53
$0.90
270.8
0.51
2.36
N/A
Fireworks (Base) logo
Meta logo
Llama 4 Maverick (Base)
1m
51
$0.39
177.7
0.42
3.24
N/A
Fireworks logo
DeepSeek logo
DeepSeek V3 (Dec '24)
128k
46
$1.31
65.0
0.82
8.51
N/A
Fireworks (Base) logo
Meta logo
Llama 4 Scout (Base)
1m
43
$0.26
162.1
0.46
3.54
N/A
Fireworks logo
Meta logo
Llama 3.3 70B
128k
41
$0.90
153.4
1.47
4.73
N/A
Fireworks logo
Meta logo
Llama 3.1 405B
128k
40
$3.00
79.0
0.88
7.21
N/A
Fireworks logo
Alibaba logo
Qwen2.5 72B
131k
40
$0.90
72.4
0.33
7.23
N/A
Fireworks logo
Meta logo
Llama 3.1 70B
128k
35
$0.90
167.6
0.39
3.38
N/A
Fireworks logo
01.AI logo
Yi-Large
32k
28
$3.00
66.9
0.39
7.86
N/A
Fireworks logo
Mistral logo
Mixtral 8x22B
65k
26
$1.20
91.6
0.33
5.79
N/A
Fireworks logo
Meta logo
Llama 3.1 8B
128k
24
$0.20
299.4
0.22
1.89
N/A

Key definitions

Context window: Maximum number of combined input & output tokens. Output tokens commonly have a significantly lower limit (varied by model).
Output Speed: Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).
Latency (Time to First Token): Time to first token received, in seconds, after API request sent. For reasoning models which share reasoning tokens, this will be the first reasoning token. For models which do not support streaming, this represents time to receive the completion.
Price: Price per token, represented as USD per million Tokens. Price is a blend of Input & Output token prices (3:1 ratio).
Output Price: Price per token generated by the model (received from the API), represented as USD per million Tokens.
Input Price: Price per token included in the request/message sent to the API, represented as USD per million Tokens.
Time period: Metrics are 'live' and are based on the past 72 hours of measurements, measurements are taken 8 times a day for single requests and 2 times per day for parallel requests.