AkashML — pricing & performance
Models served
10
of 1,094 tracked models
Mean output speed
40
tokens per second, averaged over
10 measured endpoints
10 measured endpoints
Mean latency
1.5s
time to first token, averaged over
10 measured endpoints
10 measured endpoints
Cheapest blended
$0.040
per 1M tokens at a 3:1
input:output blend
input:output blend
Output speed on AkashML
Output tokens per second for all 10 models on AkashML with published throughput telemetry, the most recent figure OpenRouter publishes for each host. Higher is better. We do not run these measurements ourselves.
All 10 of 10 models⇩
The Known Good
10 models on AkashML
Full leaderboard →
| Model | Creator | Input $/M | Output $/M | Blended | Speed | TTFT | Context |
|---|---|---|---|---|---|---|---|
DeepSeek: DeepSeek V4 Flash 0423
|
DeepSeek | $0.098 | $0.20 | $0.12 | 8 | 1.7s | 131k |
|
DeepSeek: DeepSeek V4 Flash 0731
|
DeepSeek | $0.065 | $0.18 | $0.094 | 10 | 4.0s | 1.05M |
Meta: Llama 3.3 70B Instruct
|
Meta | $0.20 | $0.52 | $0.28 | 19 | 0.8s | 131k |
OpenAI: gpt-oss-120b
|
OpenAI | $0.030 | $0.17 | $0.065 | 43 | 0.9s | 131k |
|
OpenAI: gpt-oss-20b
|
OpenAI | $0.020 | $0.10 | $0.040 | 29 | 1.4s | 131k |
Qwen: Qwen3.5-35B-A3B
|
Alibaba | $0.14 | $1.00 | $0.35 | 106 | 1.4s | 262k |
|
Qwen: Qwen3.6 35B A3B
|
Alibaba | $0.10 | $0.90 | $0.30 | 69 | 0.7s | 262k |
|
Qwen: Qwen3.8 27B
|
Alibaba | $0.25 | $2.20 | $0.74 | 51 | 1.4s | 262k |
Z.ai: GLM 5.2
|
Z.ai | $0.77 | $2.42 | $1.18 | 27 | 0.6s | 1.05M |
|
Z.ai: GLM 5.3
|
Z.ai | $1.17 | $3.96 | $1.87 | 35 | 1.9s | 1.05M |
The Known Good
Source. Prices, endpoint availability and throughput for AkashML are ingested from OpenRouter;
model metadata is enriched from models.dev. A blank cell means the figure was not published for that endpoint —
it never means zero. Blended price is (3 × input + output) ÷ 4, the same 3:1 blend used everywhere on this
site; see methodology. Quality scores are not provider-specific:
a model scores the same wherever it is hosted, so quality lives on the
model pages.