Mistral Medium 3.5
Mistral AI · dense · 127.7B parameters · 262,144 context
Parameters
127.7B
Context Window
256K tokens
Architecture
Dense
Best GPU
B200 SXM
Cheapest API
$7.50/M
Intelligence Brief
Mistral Medium 3.5 is a 127.7B parameter DENSE model from Mistral AI, featuring Grouped Query Attention (GQA) with 88 layers and 12,288 hidden dimensions. With a 262,144 token context window, it supports vision, multilingual. The most cost-effective API deployment is via openrouter at $7.50/M output tokens. For self-hosted inference, B200 SXM delivers optimal throughput at $4261/month.
Provider pricing
1 provider · canonical: openrouter| Provider | Input $/M | Output $/M ▲ | Notes |
|---|---|---|---|
| openroutercanonical | $1.50 | $7.50 | cheapest input · cheapest output |
Prices update via the nightly pricing cron + admin approvals at /admin/ingest-queue. The leaderboard's Input/Output cells show the canonical rate above; this table shows the full spread.
Recent changes
Loading…
Related models
5 suggestions- Mistral 7BMistral · 7.3B$0.070/M out
- Mistral Medium 3Mistral · 70B$2.00/M out
- Mistral Small 4Mistral · 6.6B$0.600/M out
- Mistral Large 2411Mistral Large · 123B$6.00/M out
- Mistral Large 2Mistral Large · 123B$2.50/M out
Picks: same family first, then same vendor within ±2× params, then top tag-overlap matches. Price shown is the cheapest Output $/M across providers — the row's page shows the canonical anchor.
Architecture Details
Memory Requirements
BF16 Weights
237.9 GB
FP8 Weights
118.9 GB
INT4 Weights
59.5 GB
GPU Compatibility Matrix
Mistral Medium 3.5 is compatible with 21% of GPU configurations across 41 GPUs at 3 precision levels.
GPU Recommendations
FP8 · 1 GPU · tensorrt-llm
100/100
score
Throughput
280.0 tok/s
Latency (ITL)
3.6ms
Est. TTFT
1ms
Cost/Month
$4261
Cost/M Tokens
$5.79
FP8 · 1 GPU · tensorrt-llm
100/100
score
Throughput
280.0 tok/s
Latency (ITL)
3.6ms
Est. TTFT
1ms
Cost/Month
$4271
Cost/M Tokens
$5.80
FP8 · 1 GPU · tensorrt-llm
100/100
score
Throughput
280.0 tok/s
Latency (ITL)
3.6ms
Est. TTFT
1ms
Cost/Month
$6169
Cost/M Tokens
$8.38
Deployment Options
API Deployment
openrouter
$7.50/M
output tokens
Single GPU
B200 SXM
$4261/mo
Min VRAM: 119 GB
Multi-GPU
H20 x2
280.0 tok/s
TP· $1879/mo
API Pricing Comparison
| Provider | Input $/M | Output $/M | Badges |
|---|---|---|---|
| openrouter | $1.50 | $7.50 | Cheapest |
Cost Analysis
| Provider | Input $/M | Output $/M | ~Monthly Cost |
|---|---|---|---|
| openrouterBest Value | $1.50 | $7.50 | $45 |
Cost per 1,000 Requests
Short (500 tok)
$2.25
via openrouter
Medium (2K tok)
$9.00
via openrouter
Long (8K tok)
$27.00
via openrouter
Performance Estimates
Throughput by GPU
VRAM Breakdown (B200 SXM, FP8)
Precision Impact
bf16
255.4 GB
weights/GPU
fp8
127.7 GB
weights/GPU
~280.0 tok/s
int4
63.9 GB
weights/GPU
Capabilities
Features
Supported Frameworks
Supported Precisions
Where to Deploy Mistral Medium 3.5
Self-Hosted Infrastructure
Similar Models
Mistral Small 4
119.4B params · moe
Quality: 50
from $0.60/M
Mistral Medium 3
70B params · dense
Quality: 80
from $2.00/M
DBRX Base
132B params · moe
Quality: 50
from $2.25/M
DBRX Instruct
132B params · moe
Quality: 50
from $1.20/M
Mistral Large 2411
123B params · dense
Quality: 75
from $6.00/M
Frequently Asked Questions
How much VRAM does Mistral Medium 3.5 need for inference?
Mistral Medium 3.5 requires approximately 237.9 GB of VRAM at BF16 precision, 118.9 GB at FP8, or 59.5 GB at INT4 quantization. Additional VRAM is needed for KV-cache (360448 bytes per token) and activations (~0.00 GB).
What is the best GPU for Mistral Medium 3.5?
The top recommended GPU for Mistral Medium 3.5 is the B200 SXM using FP8 precision. It achieves approximately 280.0 tokens/sec at an estimated cost of $4261/month ($5.79/M tokens). Score: 100/100.
How much does Mistral Medium 3.5 inference cost?
Mistral Medium 3.5 API inference starts from $1.50/M input tokens and $7.50/M output tokens. Self-hosted inference costs depend on your GPU configuration — use our ROI calculator for a detailed breakdown.