Mistral · Mistral Medium · Mid
Mistral Medium 3 pricing and limits
Mistral's mid-sized model priced well below frontier models while targeting strong performance on coding and multimodal tasks.
| Line | Tokens | Rate / 1M | Cost |
|---|---|---|---|
| Input, uncached | 7,000 | $0.40 | $0.00280 |
| Output | 900 | $2.00 | $0.00180 |
- Input / 1M
- $0.40
- Output / 1M
- $2.00
- Cached / 1M
- n/a
- Cache write / 1M
- n/a
- Blended 3:1
- $0.80
- Context
- 128K
What drives Mistral Medium 3 cost
Input and output prices are low, with output at five times input. There is no cache-read discount in the list price, so prompt length is billed in full on every call.
Its price point makes it a candidate replacement for frontier models on many enterprise tasks; measure steps and output length per task before switching, since a cheaper model that needs more steps can erase the saving.
Cost of a typical call
Take a assistant turn with 7,000 input tokens and 900 output tokens. At Mistral Medium 3 list prices ($0.40 input, $2.00 output per 1M) one call costs $0.00460, and 300,000 calls a month come to $1,380. The list price has no discounted cache-read rate, so every input token is billed at the full rate on every call.
Blended at a 3:1 input to output ratio, Mistral Medium 3 costs $0.80 per 1M tokens. Routing the same call to the fallback in the policy below, Mistral Small 3.1, costs $0.00097, which is what a tenant pays you in COGS once they cross the hard cap and PALM SOUQ downgrades instead of blocking.
Cost comparison
Mistral Medium 3 against three alternatives
The same assistant turn priced on Mistral Large 2, GPT-5 mini, Gemini 2.5 Flash. Move the sliders to match your traffic; models without a cache-read rate bill every input token in full.
Benchmark quality per task before switching. A cheaper model that needs more steps or longer outputs can cost more per finished task than its rate suggests, which is why PALM SOUQ reports cost per feature, not only per token.
Mistral Medium 3 this guide
$0.00460 per call
$1,150 / mo
baseline
Mistral Large 2
$0.0194 per call
$4,850 / mo
+322% vs Mistral Medium 3
GPT-5 mini
$0.00355 per call
$888 / mo
-23% vs Mistral Medium 3
Gemini 2.5 Flash
$0.00435 per call
$1,088 / mo
-5% vs Mistral Medium 3
List prices from the PALM SOUQ LLM Price Index, computed per call as uncached input + cache reads + output. Reasoning tokens count as output on reasoning models.
Rate limits by usage tier
Workspace level rate limits by tier on la Plateforme, with tokens per minute and per month ceilings. Cloud marketplace deployments follow the host cloud's quotas.
Exact limits vary by account tier and change over time, so we do not reprint them here. Check the limits page for your own account in the Mistral documentation at docs.mistral.ai. PALM SOUQ enforces your own per-tenant limits in front of the provider's, so a single customer cannot consume the shared allowance that every other tenant depends on.
Circuit-breaker best practices
Use token velocity and per-tenant monthly caps; the low rate means loops are cheap per call but long-running ones still add up.
Apply the same breaker set when it is used as a fallback for Mistral Large.
When a breaker trips, the gateway answers in under 50 ms with HTTP 429 and X-PalmSouq-Reason: circuit_breaker_triggered, before the request reaches Mistral. Your agent framework sees a normal rate-limit response and stops, and the audit log records which heuristic fired and for which tenant.
# palmsouq.policy.yaml
policy: mistral-medium-3-pro
match:
model: mistral-medium-3
tier: pro # from X-PalmSouq-Tier
budgets:
monthly_usd: 100 # per tenant
soft_cap_pct: 80 # webhook to your app
on_hard_cap: fallback # or: block (HTTP 429)
fallback:
- mistral-small-3.1
breakers:
tool_thrashing: { identical_calls: 3, window_steps: 8 }
token_velocity: { max_tokens_per_s: 50000, window_s: 10 }
semantic_echo: { repeats: 3 }
max_steps: 15- budgetsPer-tenant monthly dollar cap with an 80% soft-cap webhook, keyed by X-PalmSouq-Tenant.
- fallbackAt the hard cap, requests reroute to Mistral Small 3.1 ($0.15 blended per 1M) instead of failing.
- breakersAll four loop heuristics on, evaluated in the request path with under 5 ms added latency.
Where it fits
Enterprise assistants, document processing and coding help where frontier pricing is not justified.
Provider documentation: docs.mistral.ai
Mistral Medium 3 pricing questions
How much does Mistral Medium 3 cost per million tokens?
At list price Mistral Medium 3 costs $0.40 per 1M input tokens and $2.00 per 1M output tokens. Prices were last reviewed in the PALM SOUQ LLM Price Index and are for the standard tier and shortest context band.
Does Mistral Medium 3 support prompt caching?
The Mistral list price for Mistral Medium 3 has no discounted cache-read rate, so PALM SOUQ meters every input token at $0.40 per 1M. Shortening repeated context and caching full responses in your own application are the main savings levers.
What are the Mistral Medium 3 rate limits?
Workspace level rate limits by tier on la Plateforme, with tokens per minute and per month ceilings. Cloud marketplace deployments follow the host cloud's quotas. For your account's exact numbers, see the Mistral documentation at docs.mistral.ai.
How do I stop runaway Mistral Medium 3 agent loops?
Use token velocity and per-tenant monthly caps; the low rate means loops are cheap per call but long-running ones still add up. The sample policy on this page enables all four PALM SOUQ heuristics with a max step cap of 15 and falls back to Mistral Small 3.1 at the hard cap.
Is there a batch discount for Mistral Medium 3?
50% off through the Mistral batch API. Batch traffic suits evaluations, backfills and nightly enrichment jobs that do not need a real-time answer.
Govern Mistral Medium 3 spend per customer.
Point your Mistral client at gateway.palmsouqltd.com, pass X-PalmSouq-Tenant, and every Mistral Medium 3 token is attributed, capped and protected by circuit breakers.
14-day free production trial · 100k requests included · no card required